Agent skill

Spectre Validate

by joenandez in joenandez/spectre

Validate whether completed implementation delivers its scope/task requirements — dispatch independent @spectreanalyst checks for definition, connection, and reachability, then write one actionable…

MITAuto-check passed

Install Spectre Validate

skills CLI
$ npx skills add joenandez/spectre --skill spectre-validate -a claude-code

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

GitHub CLI
$ gh skill install joenandez/spectre spectre-validate --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/joenandez/spectre.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/spectre-codex/skills/spectre-validate .claude/skills/spectre-validate && 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
spectre-validate
GitHub stars
164
Token cost
~1.9k tokens
SKILL.md length
888 words
Files
1
Skills in repo
26
Repo updated
First seen
Licence
MIT

At a glance

Validate whether completed implementation delivers its scope/task requirements — dispatch independent @spectreanalyst checks for definition, connection, and reachability, then write one actionable…

  • Works in 7 steps: Self-location metadata — immediately… → Summary — overall status (Complete |… → Gap Remediation Tasks — phased by… → …
  • Standalone requirement validation
  • SKILL.md covers Core principle (load-bearing), Inputs, Working Set (late-bound — read… and Method / guardrails, plus 3 more sections
  • Calls git

What it does

Spectre Validate is an agent skill from joenandez/spectre. Validate whether completed implementation delivers its scope/task requirements — dispatch independent @spectreanalyst checks for definition, connection, and reachability, then write one actionable validationgaps.md. Trigger for standalone requirement validation or from Execute before observed Proof. Do NOT trigger to fix findings (spectre-fix) or as a substitute for observed Proof.

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.

The repository describes itself as: Spectre is a battle-tested 6 stage agentic engineering workflow for building complex, production-ready features. You stay in the loop for the decisions that matter. Your agents…. The licence is MIT.

When your agent uses it

  • Standalone requirement validation
  • From Execute before observed Proof
  • Fix findings (spectre-fix)
  • As a substitute for observed Proof

Example prompts

  • “/spectre-validate”

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. Self-location metadata — immediately below the title: Feature: and Feature Root: .spectre/features/.
  2. Summary — overall status (Complete | Needs Work | Significant Gaps) · {X of Y} delivered · gap count · scope-creep count · supplied…
  3. Gap Remediation Tasks — phased by priority (Critical / Medium / Low). Each gap: requirement · current state · gap; each action carries…
  4. Scope Creep Review — items beyond scope (Keep+document / Remove / Discuss) with evidence.
  5. Validation Coverage table — Area · Status · Definition · Usage · Render Chain.
  6. Dead Computations Found table — Variable · File · Computed By · Should Be Consumed By.
  7. Old Code Paths Still Active table — Old Path · Location · Replaced By · Impact.

What it can do on your machine

Read from SKILL.md and the folder at commit 931c9e8. 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

    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

Spectre Validate loads about 1.9k tokens when it runs. Until then it costs about 101 tokens; SKILL.md has 888 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~101
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 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 joenandez/spectre at commit 931c9e8, republished under its MIT licence (© joenandez). 888 words, ~1,877 tokens.

Download SKILL.mdSave it as .claude/skills/spectre-validate/SKILL.md (or your agent's skills folder).
name
spectre-validate
description
Validate whether completed implementation delivers its scope/task requirements — dispatch independent @spectre_analyst checks for definition, connection, and reachability, then write one actionable validation_gaps.md. Trigger for standalone requirement validation or from Execute before observed Proof. Do NOT trigger to fix findings (spectre-fix) or as a substitute for observed Proof.
user-invocable
true

validate

Post-implementation requirement validation. Verify the built work against scope docs and tasks.json slices, dispatch parallel validators per area, and produce one actionable gap-remediation document. Validation only — never fix.

Core principle (load-bearing)

Definition ≠ Connection ≠ Reachability. A requirement is only delivered if all three hold:

  1. Defined — code exists in a file.
  2. Connected — it is imported/called by other code.
  3. Reachable — a user action can trigger the code path.

Level 1 without 2/3 is dead code that happens to match the description — not complete. Always trace past "X exists" → "X is called by Z at file:line" → "Z fires when the user does W."

Inputs

  • $ARGUMENTS — explicit feature name/root or descendant artifact, scope docs, an explicit plan or fix/bug report, and/or tasks.json to validate against (paths, or "use thread context"), plus optional execution_state.md focus evidence, an immutable BASE_SHA/HEAD_SHA/DIFF_SHA256 candidate tuple, and --orchestrated when a parent workflow owns remediation and the next step. Requirement context is REQUIRED. If absent, ask for a scope, plan, fix/bug report, task detail JSON path, or "use thread context" — then wait.
  • Treat the supplied plan, fix/bug report, or other explicit source as authoritative. Otherwise use task acceptance slices, scope docs, then usable thread context. Read the authoritative source in full. For tasks.json, extract validation slices from phases[] (parent title/description plus child acceptance criteria/context); do not use execute.md as the validation source except to locate the JSON path.

Working Set (late-bound — read at run-time, never inline)

  • Reuse a managed FEATURE_ROOT only when explicit/current-thread evidence ties it to this work (physical directory wins; never branch/recency/lifecycle/scans); distinct work ignores ambient roots. Otherwise, including on collision, standalone MUST first load and follow Skill(spectre-feature-root) through DONE; orchestrated calls escalate. Keep writes beneath it and pass it unchanged.
  • Repair stale feature/root metadata in artifacts this workflow touches.
  • VALIDATION_REPORT = {FEATURE_ROOT}/validation/validation_gaps.md; create its parent directory.
  • Read the actual code, diff, and explicitly supplied requirement artifacts just-in-time.

Method / guardrails

  • Chunk into 1–8 validation areas: one per parent task in tasks.json; derive areas from plan outcomes/workstreams when tasks are absent; else use one per "In Scope" item in scope.md, then one per discussed feature. Keep at least one independent analyst and do not fabricate or pad areas to meet a minimum. Merge small, split large.
  • Execution state is derivative: execution_state.md is evidence/focus only for cross-wave wiring, scope-creep, and dead-computation checks; it is never requirements or acceptance authority.
  • Dispatch one @spectre_analyst per area, ALL in parallel in a single message. Brief each with: the area, its source requirement (exact text), expected deliverables, branch, and the status enum + evidence + reachability rules below. Subagents return compressed findings in-thread — no per-area files.
  • Status enum (each requirement): ✅ Delivered (defined AND connected AND reachable) · ⚠️ Partial (exists, broken/missing connection) · 🔌 Dead Code (exists, zero usage sites) · ❌ Missing (does not exist).
  • Evidence rule (YOU MUST): every requirement cites both a definition site (file:line) and a usage site (file:line). Definition only, no usage → status is ⚠️ or 🔌, never ✅.
  • Delivered evidence stays strict: every Delivered result still needs definition plus usage/reachability evidence.
  • Reachability + render-backward trace: for UI features, trace from the final render backward — JSX ← variable ← source (hook/prop/computed) ← user action. A broken link anywhere = ⚠️. Grep for usage not just definition (fnName(, <Component, useHook(, prop={).
  • Dead-computation + old-path audit: flag any computed value that never reaches a render (🔌), and any old code path still active after new code replaced it (⚠️, new path bypassed / duplicate data source).
  • Note scope creep — anything built beyond the requirements.
  • Final wiring check before any ✅: consumer connected? render chain unbroken? old path removed? no orphaned computation? single data source? Any failure downgrades ✅ → ⚠️ and adds a gap task.
  • No fixes. Report and hand back.
  • Candidate pin. When any candidate field is supplied, require all three; recompute the canonical hash using git diff --binary --full-index --no-ext-diff --no-color --no-renames before dispatch and after report creation. Record the unchanged tuple in the report; otherwise return a stale-candidate result marked incomplete. Execute always supplies all three fields.
Show full SKILL.md (243 more words)Show less

Outputs + DONE

Write VALIDATION_REPORT. Required sections:

VALIDATION_REPORT begins with its title followed immediately by the Feature/Feature Root metadata below.

  1. Self-location metadata — immediately below the title: Feature: <feature-name> and Feature Root: .spectre/features/<feature-name>.
  2. Summary — overall status (Complete | Needs Work | Significant Gaps) · {X of Y} delivered · gap count · scope-creep count · supplied candidate tuple.
  3. Gap Remediation Tasks — phased by priority (Critical / Medium / Low). Each gap: requirement · current state · gap; each action carries Produces / Consumed by / Replaces and verifiable check-box outcomes.
  4. Scope Creep Review — items beyond scope (Keep+document / Remove / Discuss) with evidence.
  5. Validation Coverage table — Area · Status · Definition · Usage · Render Chain.
  6. Dead Computations Found table — Variable · File · Computed By · Should Be Consumed By.
  7. Old Code Paths Still Active table — Old Path · Location · Replaced By · Impact.

DONE when: validation_gaps.md exists with all six numbered sections plus self-location metadata; every area has a status from the enum with definition+usage evidence; any supplied candidate tuple is recorded and unchanged; gaps are actionable (Produces/Consumed-by/Replaces); scope creep noted; no code was modified.

Handoff

Return status/delivered/gap/scope-creep/findings/report; --orchestrated: summary/gaps only.

HandoffDetails
🧭 Current phaseDone
📦 What was just doneResult
▶️ Proposed next stepRender resolved action.

Standalone Needs Work/Significant Gaps → $spectre:spectre-fix then revalidate; Standalone Complete → $spectre:spectre-execute for observed acceptance. Boundary ambiguity → Plan; flow/state/copy ambiguity → UX; pause → Handoff.

Escalate-If

  • No scope/tasks docs and no usable thread context → ask before dispatching.
  • A gap implies a scope/requirements change rather than a wiring defect → surface to the user; do not silently re-scope.

© joenandez, MIT. 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 plugins/spectre-codex/skills/spectre-validate of joenandez/spectre.

Open the folder on GitHubat commit 931c9e8

Compare with similar skills

Spectre Validate 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.

Spectre Validate compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Spectre Validate this skilljoenandez/spectre164—~1.9kAutomated safety check: PassMIT
Verification Before Completionforyourhealth111-pixel/Vibe-Skills3.6k—~1.1kAutomated safety check: PassApache-2.0
Verification Before Completionfarm-fe/farm5.6k46 repos~1kAutomated safety check: PassMIT
Verification Before CompletionjnMetaCode/superpowers-zh8.3k—~443Automated safety check: PassMIT
Complete Partial PRpydantic/pydantic-ai20k—~2.4kAutomated safety check: PassMIT
Deliverable CompletionHKUDS/OpenSpace7.8k—~731Automated safety check: PassMIT

Similar skills

  • Verification Before Completion

    foryourhealth111-pixel/Vibe-Skills

    Completion-evidence route used before claiming work is complete, fixed, passing, committed, or PR-ready.

    3.6k GitHub stars~1.1k tokensUpdated 1 mo ago
    Agent WorkflowsAuto-check passed
  • A skill your agent uses when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any…

    5.6k GitHub starsUsed in 46 repos~1k tokens
    Agent WorkflowsAuto-check passed
  • Verification Before Completion

    jnMetaCode/superpowers-zh

    Chinese-language rule that bars an agent from claiming work is done, fixed or passing until it has run a verification command and read the output.

    8.3k GitHub stars~443 tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Complete Partial PR

    pydantic/pydantic-ai

    Official

    Evaluate and complete an issue or PR where the submitted patch fixes only a narrow symptom of the reported pain point.

    20k GitHub stars~2.4k tokensUpdated today
    Auto-check passed
  • Deliverable Completion

    HKUDS/OpenSpace

    Clarifies that file creation tasks are complete when the deliverable is successfully written—no submission step required

    7.8k GitHub stars~731 tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed
  • Scout Completion

    kunchenguid/firstmate

    Load when a scout reports completion, presents a visual artifact for iteration, or is being considered for promotion to implementation.

    7.7k GitHub stars~349 tokensUpdated yesterday
    Auto-check passed

More from joenandez/spectre

All 26 skills in this repo
  • Verify Spectre

    joenandez/spectre

    A skill your agent uses when validating that the Spectre plugin itself still works after ANY change to this repo — skills, prompts, agents, hooks, the CLI, the Codex mirror, or a release.

    164 GitHub stars~1.5k tokensUpdated 7 days ago
    Auto-check passed
  • A skill your agent uses when modifying the Codex SPECTRE install flow, SessionStart continuity, project skill syncing, registry injection, or Codex-specific runtime files.

    164 GitHub stars~2.2k tokensUpdated 7 days ago
    Auto-check passed
  • Spectre Handoff

    joenandez/spectre

    Save a quiet, branch-keyed session snapshot with continuity, active work, and optional task/todo state for automatic resume.

    164 GitHub stars~1.3k tokensUpdated 7 days ago
    Auto-check passed
  • Spectre Learn

    joenandez/spectre

    User-invoked front door for durable capture. An agent skill from joenandez/spectre.

    164 GitHub stars~1k tokensUpdated 7 days ago
    Auto-check passed
  • Spectre Rewrite Skill

    joenandez/spectre

    Rewrite a verbose SPECTRE skill into a compact, contract-form Spectre skill proposal — clear on WHAT, silent on HOW — grounded in the spectre-light distillation and freedom-tier budget.

    164 GitHub stars~3k tokensUpdated 7 days ago
    Auto-check passed
  • Spectre Rewrite Subagent

    joenandez/spectre

    Rewrite Spectre subagent definitions into compact, contract-driven agents with clear routing, least-privilege tools, output schemas, and an auditable rewrite trail.

    164 GitHub stars~1.8k tokensUpdated 7 days ago
    Auto-check passed

Questions about Spectre Validate

What does Spectre Validate do?

Validate whether completed implementation delivers its scope/task requirements — dispatch independent @spectreanalyst checks for definition, connection, and reachability, then write one actionable…. Spectre Validate is an agent skill from joenandez/spectre.md.

When should I use Spectre Validate?

Spectre Validate fits situations like: standalone requirement validation; from Execute before observed Proof; fix findings (spectre-fix); as a substitute for observed Proof.

How do I install Spectre Validate in Claude Code?

Run `npx skills add joenandez/spectre --skill spectre-validate -a claude-code`. Or copy the skill folder (plugins/spectre-codex/skills/spectre-validate in joenandez/spectre) into .claude/skills/spectre-validate in your project. Claude Code loads it when a task matches its description.

How do I install Spectre Validate in Codex?

Run `npx skills add joenandez/spectre --skill spectre-validate -a codex`. Or copy the skill folder (plugins/spectre-codex/skills/spectre-validate in joenandez/spectre) into .agents/skills/spectre-validate in your project. Codex loads it when a task matches its description.

Can I use Spectre Validate 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 joenandez/spectre --skill spectre-validate -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/spectre-validate, .gemini/skills/spectre-validate, .github/skills/spectre-validate and .opencode/skills/spectre-validate in your project.

What does Spectre Validate need to run?

Going by SKILL.md and its folder, Spectre Validate needs the command-line tools its instructions call (git).

Does Spectre Validate 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 Spectre Validate 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 Spectre Validate use?

Spectre Validate 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 Spectre Validate use?

About 1.9k tokens (SKILL.md is roughly 7.5k 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 Spectre Validate?

Skills that share tags, products or a category with Spectre Validate: Verification Before Completion (foryourhealth111-pixel/Vibe-Skills, 3.6k stars), Verification Before Completion (farm-fe/farm, 5.6k stars), Verification Before Completion (jnMetaCode/superpowers-zh, 8.3k stars) and Complete Partial PR (pydantic/pydantic-ai, 20k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Spectre Validate?

joenandez (a GitHub user) maintains it in joenandez/spectre, which has 164 GitHub stars. The repository holds 26 skills in this directory. The repository was last updated on October 2, 2026.

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