acc-crit

acc-crit is a skill for Claude Code from stellarshenson/claude-code-plugins. It costs 79 tokens per session (1,464 once invoked), scanned A, original, MIT.

A workflow for adding, closing, rejecting, relating, or checking acceptance criteria in a project's tracked criteria document. Acceptance criteria are specific conditions used to decide whether work is complete.

In plain words
What is it for?
Use it to manage criteria, including permanent category-based IDs, author handles, test hints, tags, states, and relationships.
Why use it?
It keeps criteria consistently identified, authored, related, and updated through the project's command-line tools.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions .claude-plugin.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the project-management plugin — 5 skills, 4 commands, 2 hooks shipped together

Good fit Use it to manage criteria, including permanent category-based IDs, author handles, test hints, tags, states, and relationships.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add stellarshenson/claude-code-plugins
Claude Code
/plugin install project-management

Made for: Claude Code.

Or install project-management, the plugin that ships this one along with the rest of its 5 skills, 4 commands, 2 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for acc-crit

README.md
[![agentmods](https://agentmods.dev/badge/skills/stellarshenson/claude-code-plugins/acc-crit/github.svg)](https://agentmods.dev/skills/stellarshenson/claude-code-plugins/acc-crit)
Your own site
<a href="https://agentmods.dev/skills/stellarshenson/claude-code-plugins/acc-crit"><img src="https://agentmods.dev/badge/skills/stellarshenson/claude-code-plugins/acc-crit/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for acc-crit

Your own site · 80×15
<a href="https://agentmods.dev/skills/stellarshenson/claude-code-plugins/acc-crit"><img src="https://agentmods.dev/badge/skills/stellarshenson/claude-code-plugins/acc-crit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,464 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00079 $0.01464
Opus 5.5 $0.00032 $0.00586
Sonnet 5.5 $0.00016 $0.00293
Haiku 4.5 $0.00008 $0.00146

Measured 16d ago against content hash 115d54a33068, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

acc-crit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 16d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/project-management/skills/acc-crit/SKILL.md · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Acceptance Criteria

Read the project-management skill first - it is the single source of truth for the id scheme, the line format, the three states, authoring, relations and the pm-tools command surface. Do NOT duplicate its content here. Its skills/project-management/references/acceptance-criteria.md wins on any conflict about criteria specifically.

Toolchain gate (MANDATORY - run before anything else)

Run this first, every session, before any other work. The upgrade always runs; a version mismatch blocks.

python3 -m pip install --user --upgrade stellars-claude-code-plugins 2>&1 | tail -1
LIB=$(python3 -c "import importlib.metadata as m;print(m.version('stellars-claude-code-plugins'))" 2>/dev/null) || { echo "FATAL: toolkit unavailable"; exit 1; }
PLUG=$(grep -m1 '"version"' "${CLAUDE_PLUGIN_ROOT}/.claude-plugin/plugin.json" 2>/dev/null | cut -d'"' -f4)
OLDER=$(printf '%s\n%s\n' "$LIB" "$PLUG" | sort -V | head -1)
[ -n "$PLUG" ] && [ "$LIB" != "$PLUG" ] && [ "$OLDER" = "$LIB" ] && { echo "STALE: library $LIB older than plugin $PLUG - refusing to run on an outdated CLI; re-run the upgrade"; exit 1; }
echo "toolkit $LIB"

Run the CLI without touching the caller's project. The gate above puts it on PATH, so the bare command name is the whole invocation. uv run instead resolves whatever project the working directory sits in and writes uv.lock and .venv into it, so if you reach for uv pass --no-project (uv run --no-project <cli> ...) - it skips project discovery, leaves the tree untouched and still finds the same PATH binary. --no-sync and --frozen are not substitutes; both still create .venv.

Both branches exit non-zero and neither is advisory: an absent library (FATAL) and a version mismatch (STALE). A mismatch means the CLI is not the one this file was written against, so its documented flags and rules are unverified. Report the line and stop; do not work around it.

What to do

  1. Read the project-management skill, then skills/project-management/references/acceptance-criteria.md
  2. Ask for the author handle once and reuse it - every write takes --author @xx, and the handle must be on the roster (pm-tools author FILE --handle @xx --name "Full Name") before it can write
  3. Find the store before creating one - ls docs/acc-crit*.md; one consolidated doc per project is the default, a scoped file only when the user asks. Never a file per criterion
  4. Lock what you work on, release when you stop - default, unasked: pm-tools lock FILE --id ID --author @xx writes its lock: line before your first write to an item, unlock when you are done with it; ask before working on an item locked by someone else. The lock never blocks a write, expired locks clear themselves on the next write, and pm-tools unlock FILE --author @xx --id ID clears one at will; taking or clearing an active lock held by another handle is a transfer - lock and unlock print TRANSFER: ACC-AUTH-1 was locked by @yy until <stamp> - you are taking it over; ask @yy and proceed, and a takeover with no --note records taken over from @yy. report, list, search and refs print N item(s) currently worked on: ACC-AUTH-1 by @xx until <stamp> on stderr before the table - read it before choosing what to pick up
  5. One assertion per item - a criterion needing "and" is two criteria; edge cases are their own items, enumerated across the whole fanout. Write every field laconic - one reading only, nothing past the mechanism, the site, the numbers and the harm, and numbers, ids and paths copied exactly (SKILL.md, Writing the text)
  6. Rate it yourself, as you file it - --importance CRITICAL|HIGH|MEDIUM|LOW, read off the assertion against the rubric in skills/project-management/references/acceptance-criteria.md. Never ask the user for the level and never leave it unset; add refuses one and check errors on one
  7. Write through pm-tools, never by hand - add assigns the next id and the log line; hand-editing loses both
  8. Record the mechanism once the approach is decided - pm-tools mechanism FILE --id ID --text "<how it is made true>" --author @xx, or add --mechanism when it is known at filing time. Changing approach OVERRIDES: the new record goes above the old one and the old one stays, so the approach that was abandoned is still readable. --update only rewords the newest record
  9. A criterion that replaces an earlier item is a new item - never rewrite the old one to say the new thing. add the new criterion, then pm-tools relate FILE --id NEW --overrides OLD --author @xx: overrides: lands on the new item and overridden-by: on the old one, in the defects file when the old item is a defect. Reject the old item if it was still open. A link that closes a cycle holds and relate removes the older one (SKILL.md, Relations)
  10. Fill the --test hint and --test-tags as you add (tags are written upper-case - UNIT, E2E), and never tag a test that does not exist
  11. Closing needs proof - close refuses without --evidence: one line saying the test: line was run and what it showed. Run it first; do not close a criterion on the strength of the code looking right
  12. Gate - pm-tools check docs after the session; exit 0 with no errors is the bar

Read the full file on GitHub · 42 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 16d ago Changed · +1 lines 115d54a33068
  2. 23d ago Changed · +30 tokens per session a91af2395bc4
  3. 25d ago Changed df56c95873f8
  4. 1mo ago Changed 3f2f70200c8d
  5. 1mo ago First seen · 41 lines · 49 tokens per session scan A 834cea9652e8

Subscribe to this mod's changes

acc-crit is a skill published in the GitHub repository stellarshenson/claude-code-plugins (4 stars, last pushed today), licensed MIT. It adds 79 tokens to every session and 1,464 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.