Skip to content
Skills Library
AI & agents · Review & audit

validate-context

/validate-context

Audit CFD context integrity — CLAUDE.md size/shape, status freshness, ADR index sync and completeness, conventions, language, token annotations — and report PASS/WARN/FAIL with proposed fixes. Use when asked to audit CFD context, or periodically before a release.

Licence
MIT
Last repository change
10.10.2026

Assessment and security

82/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content17 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 10.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Audit CFD context integrity — CLAUDE.md size/shape, status freshness, ADR index sync and completeness, conventions, language, token annotations — and report PASS/WARN/FAIL with proposed fixes. Use when asked to audit CFD context, or periodically before a release.

Original author description

Audit CFD context integrity — CLAUDE.md size/shape, status freshness, ADR index sync and completeness, conventions, language, token annotations — and report PASS/WARN/FAIL with proposed fixes. Use when asked to audit CFD context, or periodically before a release.

How it works

SKILL.md has no level-two headings · 41 lines, 324 words

SKILL.md preview
---
name: validate-context
description: "Audit CFD context integrity — CLAUDE.md size/shape, status freshness, ADR index sync and completeness, conventions, language, token annotations — and report PASS/WARN/FAIL with proposed fixes. Use when asked to audit CFD context, or periodically before a release."
metadata:
  token-budget: "~2,000–5,000"
---

Validate the CFD context in this project. Check, in order, and output a
checklist with PASS / WARN / FAIL per item:

1. **`CLAUDE.md` size**: WARN if >150 lines, FAIL if >300.
2. **`CLAUDE.md` content shape**: should be an index of markdown
   LINKS to `docs/*.md`, not paragraphs of prose. FAIL if any section
   has >10 lines of inlined content where a link would do. WARN on
   `@docs/...` references — in Claude Code those are EAGER imports
   that load the whole file every session, defeating the index; use
   plain links plus `/session-start` for orientation reads.
3. **`docs/CURRENT_STATUS.md` freshness**: run
   `git log -1 --format=%ar -- docs/CURRENT_STATUS.md`. WARN if older
   than 2 working days; FAIL if older than 1 week.
4. **ADR index integrity**: every file matching
   `docs/decisions/[0-9]*.md` must be listed in
   `docs/decisions/_index.md`, and every row in `_index.md` must point to
   a file that exists. FAIL on mismatch. FAIL also if two
   date-prefixed ADRs (`YYYY-MM-DD-<slug>.md`) share a slug; the slug
   is the ADR's identity.
5. **ADR completeness**: every ADR has Status, Context, Decision,
   Alternatives, Consequences sections. FAIL on any ADR missing one.
6. **Convention coverage**: spot-check 3 random source files. Do their
   patterns match `docs/CONVENTIONS.md`? If not, propose updates to
   CONVENTIONS.md (do not auto-apply).
7. **Language check**: any `docs/**/*.md` file written in a non-English
   language? WARN per file.
8. **Token budget annotations**: any template file or shareable ADR
   missing the top-of-file token-cost comment? WARN per file.
9. **`AGENTS.md` alias**: must exist and resolve to `CLAUDE.md`
   (`test -L AGENTS.md && [ "$(readlink AGENTS.md)" = CLAUDE.md ]`).
   WARN if missing — agents other than Claude Code will not find the
   index. FAIL
   if it is a regular file whose content differs from `CLAUDE.md` — a
   divergent copy, or a Windows checkout without `core.symlinks`, which
   turns the link into a one-line text file. Propose replacing it with
   `ln -sf CLAUDE.md AGENTS.md`.

Output the checklist and propose concrete fixes for any failures.
Do NOT auto-apply fixes — the user reviews first.

Versions

The database has no other skills named validate-context.

Database snapshot: 13.11.2026 · package version 93226147c217 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100context-report
    Analyze Claude Code context efficiency. Use when evaluating session token usage.
  • 100cfd-tools
    建筑室外风环境CFD模拟计算与分析工具集。当用户请求一键计算、获取计算状态、查看/获取统计分析判断结论、重新统计分析、生成/查看报告书时触发此技能。涵盖室外风环境模拟的全流程计算、分析判断、报告书生成与查看等功能。
  • 84audit
    Audit MCP server tool definitions against design quality criteria. Use when: 'audit MCP tools', 'check MCP tool descriptions', 'review MCP server quality', 'tool annotations', 'readOnlyHint missing', 'parameter descriptions missing', 'check the _meta annotations',…
  • 100managing-claude-context
    Sets up and maintains the project-context files Claude Code loads — root and nested `CLAUDE.md`, path-scoped rules in `.claude/rules/`, and a decision-record index — so a repository's conventions reach the model without bloating every session. Use when a repository has no…
  • 100validate-cc-md
    Validates CLAUDE.md files for size, imports, and structure. Use when user asks to "check my CLAUDE.md", "why is my CLAUDE.md too long", "validate imports", or "fix CLAUDE.md errors". Checks file size limits (30KB warning, 50KB error), @import directives, frontmatter in…
  • 100usage-audit
    Audit a Claude Code setup for token waste and context bloat. Checks MCP servers, CLAUDE.md, skills, and settings against bloat filters. Triggers on: "audit my context", "usage audit", "token audit", "context bloat". NOT for codebase audits.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.