New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Ad Archive

by CorridorTech in CorridorTech/PoseCap

Hard-delete completed plan files (tasks Status:done, specs Status:shipped, PRDs Status:superseded, ADRs Status:superseded or deprecated) via git rm, leaving git history as the only ledger.

Apache-2.0Auto-check: notesDevelopment

Install Ad Archive

skills CLI
$ npx skills add CorridorTech/PoseCap --skill ad-archive -a claude-code

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

GitHub CLI
$ gh skill install CorridorTech/PoseCap ad-archive --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/CorridorTech/PoseCap.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/ad-archive .claude/skills/ad-archive && 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
ad-archive
GitHub stars
224
Token cost
~1.9k tokens
SKILL.md length
826 words
Files
1
Skills in repo
15
Repo updated
First seen
Licence
Apache-2.0

At a glance

Hard-delete completed plan files (tasks Status:done, specs Status:shipped, PRDs Status:superseded, ADRs Status:superseded or deprecated) via git rm, leaving git history as the only ledger.

  • Works in 5 steps: Discover candidates → Present the slate → Confirm + accepted-ADR absorption gate → …
  • The user wants to archive
  • SKILL.md covers Rationale, Step 1 — Discover candidates, Step 2 — Present the slate and Step 3 — Confirm +…, plus 4 more sections
  • Calls git

What it does

Ad Archive is an agent skill from CorridorTech/PoseCap. Hard-delete completed plan files (tasks Status:done, specs Status:shipped, PRDs Status:superseded, ADRs Status:superseded or deprecated) via git rm, leaving git history as the only ledger. Use when the user wants to archive, clean, prune, or sweep finished decision-records out of the working tree. No CHANGELOG.md, no archive/ subdir (both violate kit discipline). Accepted ADRs removable only when the user names them and an absorption check passes — the ADR's substance must be grep-findable in a binding doc…

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development, covering Architecture decision records, Git workflow and Changelog and release notes. It works with Git. The licence is Apache-2.0.

When your agent uses it

  • The user wants to archive
  • Sweep finished decision-records out of the working tree

Example prompts

  • “/ad-archive”

Requirements

  • Pre-approved tools (allowed-tools): Read, Glob, Grep, Bash

Workflow steps

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

  1. Discover candidates
  2. Present the slate
  3. Confirm + accepted-ADR absorption gate
  4. Execute
  5. Hand off

What it can do on your machine

Read from SKILL.md and the folder at commit 626701b. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Read
    • Glob
    • Grep
    • Bash

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git

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

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    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

Ad Archive loads about 1.9k tokens when it runs. Until then it costs about 146 tokens; SKILL.md has 826 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~146
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
    allowed-tools: Read, Glob, Grep, Bash

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 CorridorTech/PoseCap at commit 626701b, republished under its Apache-2.0 licence (© CorridorTech). 826 words, ~1,946 tokens.

Download SKILL.mdSave it as .claude/skills/ad-archive/SKILL.md (or your agent's skills folder).
name
ad-archive
description
Hard-delete completed plan files (tasks Status:done, specs Status:shipped, PRDs Status:superseded, ADRs Status:superseded or deprecated) via `git rm`, leaving git history as the only ledger. Use when the user wants to archive, clean, prune, or sweep finished decision-records out of the working tree. No `CHANGELOG.md`, no `archive/` subdir (both violate kit discipline). Accepted ADRs removable only when the user names them and an absorption check passes — the ADR's substance must be grep-findable in a binding doc (ARCHITECTURE.md / GUIDELINES.md / AGENTS.md / code).
allowed-tools
Read, Glob, Grep, Bash
summary
Hard-delete completed plan files (tasks / specs / PRDs / superseded ADRs) into git history. ADR-accepted requires absorption proof.

/ad-archive

Removes plan files whose decision-record lifecycle is over. Git history retains the content; the working tree stops paying context cost on artifacts the LLM no longer needs to scan.

Rationale

Decision-record artifacts (doc/adr/, doc/tasks/, doc/specs/, doc/product/) carry their lifecycle in a Status: frontmatter field. Once an item reaches a terminal state (task done, spec shipped, PRD superseded, ADR superseded / deprecated), keeping the file in the working tree adds tokens every time an agent globs the directory but adds zero binding force — the work has shipped, the decision has been replaced, or the rationale has been absorbed into a permanent doc.

/ad-archive removes them via git rm. Three rules anchor the design:

  • No CHANGELOG.md, no archive/ subdirectory. WORKFLOW.md Rule #2 forbids changelogs in narrative documents; an archive/ subdir still expands under directory globs and defeats the point. Git history is the ledger; /ad-commit after /ad-archive is the breadcrumb. git log --diff-filter=D -- doc/adr/ reaches it when needed.
  • Status-driven, not age-driven. A done task removed two days after completion is fine. An accepted ADR untouched for two years is not a candidate — accepted means binding.
  • Accepted ADRs require absorption proof. Nygard's tradition treats accepted ADRs as immutable. The kit's lean-tree value justifies removal only when the ADR's substance — not just its slug citation — has been moved into a binding document such that deletion loses no information. The skill enforces this by grepping the named target for the ADR's core claim before allowing git rm.

Step 1 — Discover candidates

Read frontmatter only; do not load full bodies. Build four candidate sets.

Tasks (doc/tasks/NNNN-*.md): include if Status: done. Exclude proposed, in-progress, blocked.

Specs (doc/specs/NNNN-*.md): include if Status: shipped. Exclude draft, accepted, anything superseded by SPEC-NNNN (the supersession chain is information; chain target stays).

PRDs (doc/product/PRD.md single-product or doc/product/<slug>.md multi-product): include if Status: superseded. Exclude draft, accepted.

ADRs (doc/adr/NNNN-*.md):

  • Auto-include if Status: superseded by ADR-NNNN or Status: deprecated. The supersession chain target stays.
  • Do not auto-include Status: accepted. Accepted ADRs require the explicit Step 3 absorption check below.

Legacy plan docs (prose files under doc/ that are not in adr/, tasks/, specs/, product/; e.g. doc/v0.2-cli-plan.md): present as candidates only when the user explicitly names them. Do not auto-include — these have no Status: field, so judgement is required.

Step 2 — Present the slate

Render one block per category with one line per candidate. Format:

### Tasks (done) — 3 candidates
- doc/tasks/0001-dogfood-agents-md.md      done   2026-05-08   "Apply kit to itself..."
- doc/tasks/0002-foundation-and-bootstrap.md  done   2026-05-08   "Foundation + bootstrap skill"
- doc/tasks/0029-skill-summary-frontmatter.md  done   2026-05-10   "Move skill descriptions into per-skill frontmatter"

### Specs (shipped) — 0 candidates
(none)

### PRDs (superseded) — 0 candidates
(none)

### ADRs (superseded / deprecated) — 2 candidates
- doc/adr/0019-domain-language-layer.md   superseded by ADR-0027   2026-05-10
- doc/adr/0024-failed-pattern.md           deprecated               2026-05-11

Print a one-line summary tail: Total: N candidates across M categories. Hard-delete via git rm; git history retains. No CHANGELOG or archive subdir.

If a category has zero candidates, render the heading and (none) — do not skip silently.

If the user named legacy plan docs, render them under their own ### Legacy plan docs (user-named) — N candidates block.

Show full SKILL.md (381 more words)Show less

Step 3 — Confirm + accepted-ADR absorption gate

Ask the user which categories to sweep. Default phrasing:

Proceed with all auto-included candidates? Or name a subset (e.g. "tasks only", "tasks + specs", "skip ADR-0019").

If the user wants to also remove one or more Status: accepted ADRs, run the absorption check before adding them to the removal set:

For each named accepted ADR:

  1. Ask the user where the decision has been absorbed. Required form: <file>#<section-or-anchor> or <file>:<line> (e.g. ARCHITECTURE.md#data-flow, GUIDELINES.md§Naming, src/lib/profiles.js:23).
  2. Extract two-to-four substance keywords from the ADR's title and Decision section (not the slug — substance). Example: ADR-0004 "File-based task tracking" → keywords file-based, task tracking, doc/tasks, Markdown.
  3. Grep the absorption target for each keyword. Require at least one keyword to match. (Title-derived literal keywords may collide; substance keywords are the test.)
  4. If the grep fails: refuse removal. Print:

    ADR-NNNN absorption not verified. Substance keyword <X> not found in <target>. Absorb the rationale first (edit the binding doc), then re-run /ad-archive and re-name this ADR.

  5. If the grep succeeds: add to the removal set, print the matched location as evidence.

Do not silently downgrade the check. The absorption rule is the design's load-bearing constraint — it is the reason hard-delete is safe.

Step 4 — Execute

Single git rm invocation for all confirmed files (one shell call, atomic stage):

bash
git rm doc/tasks/0001-...md doc/tasks/0002-...md doc/adr/0019-...md

Then run git status --short and print the staged deletions. Do not commit. Do not push. The user runs /ad-commit next to author the message and sign the DCO.

If git rm fails on any file (e.g. uncommitted local edits), abort the entire batch, print the error verbatim, and instruct the user to resolve the conflict before re-running.

Step 5 — Hand off

Print exactly:

N file(s) staged for deletion. Git history retains content.
Discoverability: `git log --diff-filter=D --name-only -- doc/`

Next: /ad-commit
Suggested subject: `docs: archive <N> implemented <kind(s)>`
Suggested body: list removed slugs and the binding doc(s) the substance was absorbed into (for accepted-ADR removals).

Output contract

Writes nothing to disk other than git rm staging. Never auto-commits. Refuses to remove Status: proposed | in-progress | blocked | draft | accepted artifacts except when the user explicitly names an accepted ADR and the absorption check passes. Refuses to create CHANGELOG.md or doc/<kind>/archive/ subdirectories — both violate kit discipline (Rule #2 and the lean-tree value respectively).

Next

  • /ad-commit — author the deletion commit (Conventional Commits + DCO sign-off).
  • /ad-next — re-run the state survey; removed artifacts disappear from Layer 4 / Layer 5 counts.
  • /ad-drift — confirm no narrative document still cites the removed ADRs by slug as load-bearing context (decoration-citation rule per ADR-0030 §11).

© CorridorTech, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/ad-archive of CorridorTech/PoseCap.

Open the folder on GitHubat commit 626701b

Compare with similar skills

Ad Archive 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.

Ad Archive compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Ad Archive this skillCorridorTech/PoseCap224—~1.9kAutomated safety check: NotesApache-2.0
Ad Archivealexandremendoncaalvaro/CorridorKey-Runtime7551 repos~1.6kAutomated safety check: PassCustom licence
Release Bumpjamiepine/voicebox57k—~1.1kAutomated safety check: PassMIT
Git Workflow and Versioningaddyosmani/agent-skills103k2 repos~3.5kAutomated safety check: NotesMIT
Go-Redis Release Preparationredis/go-redis22k—~1.1kAutomated safety check: PassBSD-2-Clause
pybind11 Release Preparationpybind/pybind1118k—~1.7kAutomated safety check: PassCustom licence

Similar skills

  • Ad Archive

    alexandremendoncaalvaro/CorridorKey-Runtime

    Sweep completed plan files (tasks Status:done, specs Status:shipped, PRDs Status:superseded, ADRs Status:superseded or deprecated) out of the working tree and into git history via git rm.

    755 GitHub starsUsed in 1 repo~1.6k tokens
    DevelopmentAuto-check passed

  • Release Bump

    jamiepine/voicebox

    Ends a release cycle by moving the Unreleased changelog notes under a dated version heading, bumping version files with bumpversion and tagging the commit.

    57k GitHub stars~1.1k tokensUpdated yesterday
    DevelopmentAuto-check passed

  • Git Workflow and Versioning

    addyosmani/agent-skills

    Sets git habits for every change: short-lived branches, atomic commits with descriptive messages, clean pull requests, plus versioning, tagging and changelogs for releases.

    103k GitHub starsUsed in 2 repos~3.5k tokens
    DevelopmentAuto-check: notes

  • Go-Redis Release Preparation

    redis/go-redis

    Official

    Prepares a go-redis release locally: picks the next semver, gathers merged PRs, writes the RELEASE-NOTES entry and bumps versions, without publishing.

    22k GitHub stars~1.1k tokensUpdated today
    DevelopmentAuto-check passed

  • pybind11 Release Preparation

    pybind/pybind11

    Opens the pybind11 release-preparation pull request: picking the release base, bumping the version in common.h and integrating the changelog, following docs/release.rst.

    18k GitHub stars~1.7k tokensUpdated today
    DevelopmentAuto-check passed

  • AionUi Version Bump

    iOfficeAI/AionUi

    Automates an AionUi release: checks the latest AionCore release and its artifacts, updates package.json, writes the changelog, opens a PR and tags the release.

    33k GitHub stars~2.1k tokensUpdated 29 days ago
    DevelopmentAuto-check passed

More from CorridorTech/PoseCap

All 15 skills in this repo

  • Ad Adr

    CorridorTech/PoseCap

    Draft a new ADR (Architecture Decision Record) at doc/adr/NNNN-<short-title.md, using Michael Nygard's Context/Decision/Consequences/Alternatives pattern.

    224 GitHub stars~1k tokensUpdated yesterday
    Auto-check: notes

  • Ad Architecture

    CorridorTech/PoseCap

    Generate ARCHITECTURE.md at the repo root by scanning the code first, pre-filling layers/patterns/observability/deployment from observed signals, then asking only the genuine gaps.

    224 GitHub stars~1.5k tokensUpdated yesterday
    Auto-check: notes

  • Ad Drift

    CorridorTech/PoseCap

    Read-only drift audit — compare AGENTS.md, ARCHITECTURE.md, ADR statuses, feature specs in doc/specs/, and documentation discipline against what the code actually does.

    224 GitHub stars~1.8k tokensUpdated yesterday
    Auto-check: notes

  • Ad Hooks

    CorridorTech/PoseCap

    Scaffold deterministic quality gates per WORKFLOW.md §11 — pre-commit (lint, format, secret-scan), pre-push (build, unit, integration).

    224 GitHub stars~2.4k tokensUpdated yesterday
    Auto-check: notes

  • Ad Next

    CorridorTech/PoseCap

    Survey the project's state across the six-layer artifact stack and recommend prioritized next actions, modeled on flutter doctor.

    224 GitHub stars~3.3k tokensUpdated yesterday
    Auto-check: notes

  • Ad Bootstrap

    CorridorTech/PoseCap

    Generate AGENTS.md at the repo root by scanning the codebase first, pre-filling placeholders from observed signals, and asking only the genuine gaps.

    224 GitHub stars~2.4k tokensUpdated yesterday
    Auto-check: warnings

Works with

Categories

Questions about Ad Archive

What does Ad Archive do?

Hard-delete completed plan files (tasks Status:done, specs Status:shipped, PRDs Status:superseded, ADRs Status:superseded or deprecated) via git rm, leaving git history as the only ledger. Ad Archive is an agent skill from CorridorTech/PoseCap. Hard-delete completed plan files (tasks Status:done, specs Status:shipped, PRDs Status:superseded, ADRs Status:superseded or deprecated) via git rm, leaving git history as the only ledger.

When should I use Ad Archive?

Ad Archive fits situations like: the user wants to archive; sweep finished decision-records out of the working tree.

How do I install Ad Archive in Claude Code?

Run `npx skills add CorridorTech/PoseCap --skill ad-archive -a claude-code`. Or copy the skill folder (.claude/skills/ad-archive in CorridorTech/PoseCap) into .claude/skills/ad-archive in your project. Claude Code loads it when a task matches its description.

How do I install Ad Archive in Codex?

Run `npx skills add CorridorTech/PoseCap --skill ad-archive -a codex`. Or copy the skill folder (.claude/skills/ad-archive in CorridorTech/PoseCap) into .agents/skills/ad-archive in your project. Codex loads it when a task matches its description.

Can I use Ad Archive 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 CorridorTech/PoseCap --skill ad-archive -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ad-archive, .gemini/skills/ad-archive, .github/skills/ad-archive and .opencode/skills/ad-archive in your project.

What does Ad Archive need to run?

Going by SKILL.md and its folder, Ad Archive needs the command-line tools its instructions call (git). Its frontmatter pre-approves these tools: Read, Glob, Grep, Bash.

Does Ad Archive access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Ad Archive safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Ad Archive use?

Ad Archive is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Ad Archive use?

About 1.9k tokens (SKILL.md is roughly 7.8k 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 Ad Archive?

Skills that share tags, products or a category with Ad Archive: Ad Archive (alexandremendoncaalvaro/CorridorKey-Runtime, 755 stars), Release Bump (jamiepine/voicebox, 57k stars), Git Workflow and Versioning (addyosmani/agent-skills, 103k stars) and Go-Redis Release Preparation (redis/go-redis, 22k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ad Archive?

CorridorTech (a GitHub organization) maintains it in CorridorTech/PoseCap, which has 224 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on October 7, 2026.

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