Agent skill

Complete

by aiblueprinthq in aiblueprinthq/ai-blueprint

Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval to land it locally or by pull request.

MITAuto-check passedDevelopment

Install Complete

skills CLI
$ npx skills add aiblueprinthq/ai-blueprint --skill complete -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install aiblueprinthq/ai-blueprint complete --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/aiblueprinthq/ai-blueprint.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/complete .claude/skills/complete && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
complete
GitHub stars
463
Token cost
~5.9k tokens
SKILL.md length
3,341 words
Files
2
Skills in repo
20
Repo updated
First seen
Licence
MIT

At a glance

Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval to land it locally or by pull request.

  • Works in 4 steps: final safety pass → log the work → make the work commit → …
  • Requests to finish
  • SKILL.md covers Before you start, Configured regular quality gates, Step 0 - final safety pass and Step 1 - log the work, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Complete is an agent skill from aiblueprinthq/ai-blueprint. Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval to land it locally or by pull request. Use for /complete or requests to finish, wrap up, merge, or close the current work item.

Its SKILL.md is about 5.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `reference/completion-recovery.md`).

It sits in Development, covering Pull requests. The repository describes itself as: A file-backed, spec-driven AI coding workflow framework for building real software while staying in control. The licence is MIT.

When your agent uses it

  • Requests to finish
  • Close the current work item

Example prompts

  • “/complete”

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. final safety pass
  2. log the work
  3. make the work commit
  4. land the work

What it can do on your machine

Read from SKILL.md and the folder at commit 96222b7. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Complete loads about 5.9k tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 3,341 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~73
When it runs · the whole SKILL.md, loaded when a task matches
~5.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from aiblueprinthq/ai-blueprint at commit 96222b7, republished under its MIT licence (© aiblueprinthq). 3,341 words, ~5,932 tokens.

Download SKILL.mdSave it as .claude/skills/complete/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
complete
description
Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval to land it locally or by pull request. Use for /complete or requests to finish, wrap up, merge, or close the current work item.

complete - log the finished work, make the work commit, and land it

Context reuse: Reuse any required file already loaded in project instructions or the current session. Read it again only if absent, changed, or exact current bytes or line references are needed.

First action: Before project inspection, preflight, or any other tool call, publish running to blueprint/.state/run.json using the dashboard activity contract in AGENTS.md.

Where this sits in the workflow:

/feature, /fix, or /rollback  ->  /implement  ->  [complete]  ->  next
(the spec)                         (build it)      (commit + land + log)

/implement built the feature, fix, or rollback on its branch, with optional per-step commit checkpoints. This skill closes it out: it logs the work, makes the single work-level commit, then follows git.landing. The default local-merge path squash-merges locally. The optional pull-request path pushes the work branch and opens a pull request for a later squash merge. Run it only when the work is done, reviewed, and the documented Verify command, or the fallback build and tests, passes.

Before you start

Read blueprint/config.json. A missing file means the built-in defaults apply. If the file exists but is invalid, stop and point the user to /doctor. Configuration can strengthen or shape the completion gates, but it never grants permission to commit, merge, push, deploy, publish, or take destructive action. Record git.landing; a missing value defaults to local-merge. The pull-request value applies only to this skill. Continuous Mode remains a local-only serial workflow and never pushes or opens pull requests.

Before requiring a real active spec, check for pending completion using reference/completion-recovery.md. A matching archive may mean archival was interrupted, the work commit awaits merge, or the merge already finished. Follow that phase instead of restarting logging. Missing or ambiguous evidence stops with concrete recovery steps; a reset stub never authorizes selecting new work. Recovery uses archive and Git proof, never dashboard activity as authority.

For a normal completion with no recovery in progress, continue below.

Confirm the work is actually finished: blueprint/context/current-feature.md holds a real spec, its steps are built on a branch, and Verify, or the fallback build and tests, passes. Apply the configured regular quality gates below before logging or committing. Uncommitted step work is expected because per-step checkpoints are optional; this skill commits it. Don't require the steps to be pre-committed.

Read blueprint/context/review.md when present. A missing file on an older install means no independent review has been requested. A pending, changes-requested, malformed, or stale record is always a blocker because the user already initiated that gate, even when its configured policy is manual. Complete must route a pending or missing required review through the configured execution flow below rather than treating the manual handoff as unconditional.

Configured regular quality gates

Use qualityGates.regular for this work item:

  • Audit: manual runs only when the user explicitly requests /audit; when-sensitive runs for authentication, authorization, payments, secrets, personal or user data, migrations, destructive operations, external side effects, security boundaries, or unusually broad changes; always runs for every work item.
  • Independent review: manual runs only when the user explicitly requests /audit independent current; when-sensitive requires it for the same sensitive categories as Audit; always requires it for every work item. An independent review includes a full current audit, so it satisfies a selected Audit gate instead of repeating the audit in the builder session.
  • Check: manual runs only when explicitly requested; when-behavioral runs when a done-when needs observed runtime behavior such as a click, request, CLI command, download, background job, or multi-screen flow; always runs for every work item.
  • Try guide (qualityGates.regular.tryGuide): use /check guide. manual runs only when explicitly requested; when-user-facing generates a guide when the change affects UI, navigation, copy, a public API or CLI, output, or another workflow a person directly uses; always generates one for every work item.

Apply automatic gates in this order: /check, review, then /check guide. When independent review is selected, follow the independent execution flow below and continue only after a fresh reviewer writes a current passing receipt. Otherwise run /audit current when Audit is selected. Reuse adequate evidence produced during the current work item instead of repeating it. A required gate that cannot run is a blocker. /check guide only generates instructions for human review; never claim the user performed them. P0 and P1 finding blockers remain enforced regardless of these settings.

Independent review execution

After final Verify and required Check pass, set the active spec to verified. If a selected or previously initiated independent review does not already have a current passing receipt:

  1. Use an existing current pending request and its immutable target when one is present. Otherwise show the exact product/test checkpoint candidate and verified spec. Include the spec when tracked; an intentionally ignored spec uses Audit's local Spec snapshot contract without changing visibility. Obtain explicit commit approval when the exact checkpoint does not already exist, then create or use it under the normal Git rules. Configuration, including review.independentExecution: "automatic", never grants permission to commit. A pending request without Requested execution is legacy and manual-only; never add execution fields or run a subagent against it. A local-spec-only revision may reuse the same approved product HEAD after normal spec and verification gates, with a new snapshot/request and full fresh review. Do not create an empty commit for ignored spec changes.
  2. Prepare Phase A of /audit independent current when no current request exists. Record Requested execution from review.independentExecution.
  3. For requested automatic, start the generic isolated current-runtime child from the installed project-local Audit skill, wait, and validate the normal receipt. Freeze parent product, test, spec, and config changes while it runs.
  4. For requested manual, or when automatic isolation, identity, model, completion, or access to the same local spec/snapshot is unavailable, preserve the pending request, set activity to ready, and stop with the manual fresh-session handoff.
  5. Continue Complete only with a current passing receipt whose requested and actual execution fields match the allowed review contract. Never self-review or silently skip the gate.

Step 0 - final safety pass

Before logging or committing, run a short safety pass and report blockers only:

  • active spec exists and the work is not being completed from main or master
  • the branch name uses the configured feature, fix, or rollback prefix
  • changed files are tied to the active spec, with no unrelated dirty work mixed in (dirty blueprint/context/findings.md and blueprint/context/review.md are expected review evidence)
  • the exact Verify command from AGENTS.md passed in this session, when one is declared; otherwise the build passed, and tests passed when the project has a declared test command and the change touched logic
  • any check required by qualityGates.regular has evidence, and there is a clear manual try path
  • with verification.logicTests: "required", logic changes have a configured runner and passing focused tests; otherwise completion is blocked and /tests is the next setup step
  • with verification.uiEvidence: "required", UI done-whens have direct browser evidence, including screenshots and relevant console and network checks
  • any audit or try guide required by qualityGates.regular ran before completion
  • a selected independent-review gate has a passed receipt whose target equals HEAD, whose base ref still produces the recorded merge base, whose exact spec SHA-256 still matches, whose reviewer adapter and selected model match the request, whose required Check result passed, whose receipt sections are non-empty, and whose target has no later changes except the review and findings files. Apply the same checks to any explicit receipt even when the configured policy is manual. Any mismatch is stale and blocks completion. When Spec snapshot is present, also require its exact local bytes, path, visibility, and Git conditions from Audit's reference contract to remain valid.
  • when a passing independent receipt exists, the active spec is already verified and remains byte-for-byte unchanged through archival
  • if workflow files changed, .agents and .claude stayed in sync where both adapters exist
  • no P0 or P1 finding in blueprint/context/findings.md is open or fixed. fixed still blocks on purpose: the repair exists but no review has looked at it - run /audit to close it. While the current spec is active, append the repair as a new checklist step and run /implement; do not start a separate /fix. After the repair passes its focused check and the finding is marked fixed, run /audit current to re-review the updated code, close the finding, and then retry /complete. The only waivers are accepted (the user's explicit decision in the current chat, reason recorded; never set it for them) or invalid (an /audit re-examination verdict with recorded evidence, or the user's explicit call). A missing ledger file means no findings.

Do not claim "passed", "verified", or "working" without naming the command, route, screenshot, or output that proves it. Stop before Step 1 if required evidence is missing.

After this safety pass succeeds, set the active spec's **Status:** to verified before archiving it when no independent receipt exists. With a passing independent receipt, it must already be verified; do not rewrite it after review. Rerun the required final checks either way because /complete owns the final safety pass.

Show full SKILL.md (1,657 more words)Show less

Step 1 - log the work

Follow reference/completion-recovery.md to capture the source tree and compact archive annotation before any logging edits. Preserve the exact verified spec prefix, its UTF-8 byte length and SHA-256, branch, original HEAD, and local base. Record the reference's narrow absentOptional proof before creating any optional findings/review stub on an older installation; tree absence alone is insufficient. Prepare the entire archive, including the sections below and any generated try guide, before placing it at its absent destination. An existing matching archive enters recovery; never overwrite it or append duplicate sections.

Check whether the spec is a feature, fix, or rollback. A fix is marked Type: Fix and has no build-plan number. A rollback is marked Type: Rollback and records the exact target feature, archive, commit, and parent.

  • Feature - use the verified spec's frozen **Build attempt:** and stable ID, following ../feature/reference/build-history.md. Attempt 1 keeps blueprint/history/features/NN-name.md; N > 1 uses NN-name--build-N.md. Validate the attempt against prior immutable builds and completed reversals; never increment it at completion or resume. For an older active spec without the field, derive 1 only with no prior builds, or derive the next attempt only from unambiguous prior builds and their proven completed reversals. Keep those reviewed spec bytes and the existing branch unchanged; otherwise stop. An existing matching archive and its annotation freeze the recovery destination. Check the feature off in blueprint/build-plan.md (and its parent item once all sub-items are checked). Then recompute the overview fingerprint using /overview's checkbox-normalized hash contract and replace only the existing blueprint:source-hash value. Do not regenerate or rewrite the overview body. This keeps completion progress from looking like plan drift and migrates older exact-byte fingerprints.
  • Fix - archive it to blueprint/history/fixes/name.md. A fix isn't a build-plan item, so there's nothing to check off.
  • Rollback - archive it to blueprint/history/rollbacks/YYYY-MM-DD-<exact-target-archive-stem>.md, using the spec's exact Target archive filename without .md, including any build suffix. Preserve the original completed feature archive. Create blueprint/history/rollbacks/ first if an older Blueprint installation does not have it yet. Uncheck the exact target item in blueprint/build-plan.md and its parent when applicable, then append a concise note to the target line with the rollback date and archive path. Keep the feature number stable. If the user later decides the feature is permanently abandoned rather than pending rebuild, that roadmap decision is a separate plan edit.

Archive resolved findings. If blueprint/context/findings.md holds any findings, include a ## Findings section in the prepared archive with every closed, accepted, or invalid entry at its final status (accepted entries keep their recorded reason). Prefix feature IDs with the normalized stable ID and, for N > 1, -build-N: feature 12's first F-03 becomes 12/F-03, while attempt 2 becomes 12-build-2/F-03. Derive N from the verified spec/history proof, never arbitrary filename text. Fixes and rollbacks use their archive filename as the prefix. An entry carried forward from earlier work archives with the item that resolved it; its Found line preserves where it came from. Only closed, accepted, and invalid entries are resolved for archival. A fixed entry is not resolved at any severity: never append it to the archive or remove it from the live ledger.

Archive independent review. When a current passed receipt exists, include a ## Independent review section in the prepared archive with the receipt fields, commands, safe evidence references, findings, and remaining risk from blueprint/context/review.md. Preserve the full target and base SHAs, spec hash, the original Spec snapshot field when present, base ref, builder adapter and model, requested reviewer, model, and execution, actual reviewer adapter, model, and execution, Check result, fresh-context declaration, and review time. Do not archive a stale, pending, changes-requested, or malformed record.

Validate and place the fully assembled archive first. Complete only the exact plan/overview changes above and any approved consumed-prototype cleanup below. Read the archive back and confirm that its spec, findings, and review match their inputs before resetting live evidence. Reset the active spec last.

Discard consumed prototypes. If this feature built the look from prototypes/

  • its Design reference pointed there and an early step ported prototypes/theme.css into the app - delete the prototypes/ folder now. The tokens live in the real stylesheet and the HTML mockups were always throwaway; fold the deletion into this feature's commit. Skip this if the feature didn't consume prototypes.

Then remove only the archived entries from the ledger. Entries with open, fixed, or unverified status stay in the ledger with their IDs so they are never silently dropped. A fixed P2/P3 finding does not block completion, but it must remain verbatim for a later /audit re-review. When no open, fixed, or unverified entries remain, reset the ledger to exactly this stub, and create it the same way if the file is missing (an older install):

# Findings

> **Generated file.** The findings ledger: review findings raised by `/audit`
> against the work in progress, each with a durable ID, severity (P0-P3), and
> status. `/implement` marks repaired findings `fixed`, a later `/audit` pass
> moves them to `closed`, and `/complete` refuses to land while any P0 or P1
> finding is `open` or `fixed`, then archives resolved findings with the work
> and resets this file.

_No findings recorded. `/audit` appends findings here when it finds them._

Then reset blueprint/context/review.md to exactly this stub, creating it when an older installation does not have it:

# Independent Review

> **Generated file.** Holds the active independent-review request or latest
> receipt for the current work item. `/audit independent current` prepares a
> handoff against an approved checkpoint, a fresh reviewer context completes it,
> and `/complete` refuses stale, pending, or changes-requested review state.

_No independent review requested. Run `/audit independent current` to prepare one._

Keep every unresolved entry in the ledger. Do not replace it with the empty stub while it still contains any open, fixed, or unverified finding. After archiving resolved findings, replace blueprint/context/current-feature.md with the canonical stub below. Do not paraphrase it or substitute an abbreviated "no work" stub. Before committing, read the file and confirm it exactly matches:

# Current Feature

> **Generated file.** Holds the one feature, fix, or rollback being built in this checkout. Run
> `/feature <number-or-name>` to spec a build-plan feature, or `/fix "<bug>"` for
> an ad-hoc fix. Use `/rollback <completed-feature>` to plan a safe reversal.
> Build one thing at a time in this checkout; `/complete` archives it under
> `blueprint/history/` and resets this file.

_Nothing in progress. Run `/feature`, `/fix`, or `/rollback` to start._

When no open, fixed, or unverified ledger entries remain, confirm blueprint/context/findings.md exactly matches the canonical Findings stub above. Otherwise, preserve the remaining entries without rewriting them. Confirm blueprint/context/review.md exactly matches the canonical Independent Review stub above.

Don't commit yet; the next step makes one work commit covering the code and these documentation changes. The archive is the build history.

Step 2 - make the work commit

Show the complete product and logging diff with the proposed commit message, then obtain explicit commit approval. Only then stage the reviewed branch work (any uncommitted step work plus the Step 1 logging changes) and make one conventional work commit (for example feat: <feature>, fix: <name>, or revert: roll back <feature>). Verify, or the fallback build and tests, must pass first.

Step 3 - land the work

Follow only the configured path.

Local merge

With git.landing: "local-merge":

  1. Confirm the recorded local default branch has not advanced and the final work commit is unchanged. Squash-merge into that default branch only with the user's explicit go-ahead, so the work lands as one clean commit regardless of how many checkpoints the branch carried.
  2. Verify the resulting default-branch commit, parent, archive, and full tree using reference/completion-recovery.md, then perform approved branch cleanup.
  3. Stop and ask whether to push the local default branch to its upstream. The merge approval does not count as push approval.
  4. Push the default branch only after a separate explicit yes in the current chat. If the repo has no remote or upstream, say so instead of guessing.
Pull request

With git.landing: "pull-request":

  1. Require a selected configured remote, a known remote default branch, and an available authenticated hosting CLI or API. Prefer the work branch's configured upstream remote; otherwise use the only unambiguous remote or ask the user to choose. Confirm the exact work commit and branch are unchanged. Do not fall back to a local merge or direct default-branch push when any prerequisite is missing.
  2. Show the target remote and base branch, exact branch to push, proposed pull request title, and concise body. State that the pull request must use squash merge to preserve Blueprint's one-work-item, one-default-branch-commit history.
  3. Obtain one explicit approval for the feature-branch push and pull request creation. Configuration never supplies this approval. Only after approval, push that branch with upstream tracking and open the pull request against the confirmed default branch.
  4. Verify that the pull request head is the exact final work commit and report its URL plus check status. Never merge the pull request, enable auto-merge, delete the remote branch, or push the default branch from this approval.
  5. Stop with the work ready for provider review and squash merge. If /complete is resumed after the provider reports it merged, follow reference/completion-recovery.md; verify the exact pull request and squash commit before reporting completion or offering local branch cleanup.

Then point the user at /feature, /fix, or /rollback for the next thing.

Finish with a concise How to try it note for the completed work. For a rollback, explain how to confirm the removed behavior is gone and name one unaffected regression path. If the manual path is more than a couple of steps, tell the user to run /check guide latest; that command can read the archived feature after current-feature.md is reset.

Rules

  • The work item is the unit of history: one squashed feature, fix, or rollback commit on the default branch, even if the branch carried several checkpoint commits.
  • A rollback preserves the original feature archive and adds a separate rollback archive. Never rewrite history to make the feature look as if it never existed.
  • Don't merge unfinished or failing work. The documented Verify command, or the fallback build and tests, must pass first.
  • Never merge while a P0 or P1 finding is open or fixed in the ledger. The recorded ways past the gate without code are accepted (only by the user's explicit decision, with their reason) or invalid (only from re-examination evidence or the user's explicit call); both travel into the archive, never a silent drop.
  • Never merge with a required or explicitly initiated independent review that is missing, pending, changes-requested, malformed, or stale. The user may explicitly cancel a manual review before completion, but the agent never resets or waives it on the user's behalf.
  • Merging and pushing are the user's calls. Local landing needs explicit merge approval, then separate default-branch push approval. Pull-request landing needs explicit branch-push and pull-request approval, and never includes merge approval.
  • Never treat /complete, configuration, or "looks good" as permission to push, open a pull request, enable auto-merge, or merge.
  • One item per completion. If a parent feature still has unchecked sub-features, leave the parent unchecked.

Formatting

Format the output to match the project's conventions in blueprint/context/ai-interaction.md: concise, scannable markdown, with lists for enumerations and tables for matrices rather than dense paragraphs.

© aiblueprinthq, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .agents/skills/complete of aiblueprinthq/ai-blueprint.

  • SKILL.md
  • reference/completion-recovery.md

Open the folder on GitHubat commit 96222b7

Compare with similar skills

Complete next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Complete compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Complete this skillaiblueprinthq/ai-blueprint463—~5.9kAutomated safety check: PassMIT
Finishing a Development Branchobra/superpowers297k5 repos~1.9kAutomated safety check: PassMIT
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Check PRonyx-dot-app/onyx32k2 repos~2.3kAutomated safety check: PassMIT
PR Design DocOpenHands/OpenHands90k—~2.4kAutomated safety check: PassMIT
WooCommerce Code Reviewwoocommerce/woocommerce11k3 repos~1.1kAutomated safety check: PassCustom licence

Similar skills

  • Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.

    297k GitHub starsUsed in 5 repos~1.9k tokens
    DevelopmentAuto-check passed
  • PR Babysitter

    openinterpreter/openinterpreter

    Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed
  • Check PR

    onyx-dot-app/onyx

    Checks a GitHub, GitLab, or Perforce (p4) pull request (or merge request, or shelved changelist) for unresolved review comments, failing status checks, and incomplete PR descriptions.

    32k GitHub starsUsed in 2 repos~2.3k tokens
    DevelopmentAuto-check passed
  • PR Design Doc

    OpenHands/OpenHands

    For a non-trivial pull request, write a self-contained HTML design doc under the temporary .pr/ directory and link a visibility-appropriate preview in the PR description, so maintainers grasp the…

    90k GitHub stars~2.4k tokensUpdated today
    DevelopmentAuto-check passed
  • WooCommerce Code Review

    woocommerce/woocommerce

    Reviews WooCommerce code changes against the project's standards, flagging backend PHP architecture, naming, documentation, data integrity and testing violations.

    11k GitHub starsUsed in 3 repos~1.1k tokens
    DevelopmentAuto-check passed
  • Record PR Demo

    payloadcms/payload

    A skill your agent uses when a Payload pull request needs a concise visual walkthrough for reviewers.

    45k GitHub stars~1k tokensUpdated today
    DevelopmentAuto-check passed

More from aiblueprinthq/ai-blueprint

All 20 skills in this repo
  • Adopt

    aiblueprinthq/ai-blueprint

    Adopt Blueprint into an existing brownfield codebase by surveying shipped behavior and generating plans, standards, commands, adapter choices, and visibility setup.

    463 GitHub stars~2.7k tokensUpdated today
    Auto-check passed
  • CI

    aiblueprinthq/ai-blueprint

    Set up or normalize one project Verify command and matching GitHub Actions checks while preserving existing CI, with an optional local pre-push hook.

    463 GitHub stars~2.2k tokensUpdated today
    Auto-check passed
  • Doctor

    aiblueprinthq/ai-blueprint

    Run a Blueprint health and context check covering setup, adapters, commands, visibility, plans, overview freshness, configuration, dashboard state, and workflow drift.

    463 GitHub stars~4.5k tokensUpdated today
    Auto-check: notes
  • Feature

    aiblueprinthq/ai-blueprint

    Turn the next, named, or numbered build-plan feature into a buildable current-feature.md spec with small steps and done-when criteria.

    463 GitHub stars~2.8k tokensUpdated today
    Auto-check passed
  • Onboard

    aiblueprinthq/ai-blueprint

    Onboard a fresh or early scaffold after Blueprint is overlaid by tuning commands, standards, adapters, visibility, and context loading.

    463 GitHub stars~4.5k tokensUpdated today
    Auto-check passed
  • Overview

    aiblueprinthq/ai-blueprint

    Validate and normalize project-plan.md and build-plan.md, then generate the durable project-overview.md used by agents.

    463 GitHub stars~3.8k tokensUpdated today
    Auto-check passed

Categories

Questions about Complete

What does Complete do?

Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval to land it locally or by pull request. Complete is an agent skill from aiblueprinthq/ai-blueprint. Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval to land it locally or by pull request.

When should I use Complete?

Complete fits situations like: requests to finish; close the current work item.

How do I install Complete in Claude Code?

Run `npx skills add aiblueprinthq/ai-blueprint --skill complete -a claude-code`. Or copy the skill folder (.agents/skills/complete in aiblueprinthq/ai-blueprint) into .claude/skills/complete in your project. Claude Code loads it when a task matches its description.

How do I install Complete in Codex?

Run `npx skills add aiblueprinthq/ai-blueprint --skill complete -a codex`. Or copy the skill folder (.agents/skills/complete in aiblueprinthq/ai-blueprint) into .agents/skills/complete in your project. Codex loads it when a task matches its description.

Can I use Complete in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add aiblueprinthq/ai-blueprint --skill complete -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/complete, .gemini/skills/complete, .github/skills/complete and .opencode/skills/complete in your project.

What does Complete need to run?

SKILL.md names no scripts, command-line tools or credentials: Complete is instructions for the agent only.

Does Complete access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Complete safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Complete use?

Complete is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Complete use?

About 5.9k tokens (SKILL.md is roughly 24k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Complete?

Skills that share tags, products or a category with Complete: Finishing a Development Branch (obra/superpowers, 297k stars), PR Babysitter (openinterpreter/openinterpreter, 69k stars), Check PR (onyx-dot-app/onyx, 32k stars) and PR Design Doc (OpenHands/OpenHands, 90k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Complete?

aiblueprinthq (a GitHub organization) maintains it in aiblueprinthq/ai-blueprint, which has 463 GitHub stars. The repository holds 20 skills in this directory. The repository was last updated on October 8, 2026.

Source: aiblueprinthq/ai-blueprint on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.