agentleFS
haqalizhaqaliz/assay/.claude/skills/assay-report/SKILL.md

assay-report

Skill0 starsChanged 3 days agoApache-2.0View on GitHub

Use when an Assay unit of work (bug, task, feature) is done and you want a brief, friendly, non-technical completion note saved in the Projects Brain vault to share with the team.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No checklists, no commit hashes, no file paths.

    line 49
  2. Very specific:

    Don't say a skill is "safe" when the verdict is CLEAN — say the observed run matched what it claimed, this time.

    line 54
  3. Very specific:

    Don't imply MALICIOUS without saying it was reproduced, captured behavior — never an accusation based on how a skill reads.

    line 56
  4. Distinctive:

    If the work is a slice of a bigger capability, say it's a first step.

    line 58
  5. Distinctive:

    Ask for a screenshot or short video: if one isn't on hand.

    line 67

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Assay Completion Note

A short, friendly, non-technical heads-up that a unit of work is done.

Arguments

  • type ∈ bug | task | feature
  • id = the GitHub issue number, or the slug used at begin time

Usage: /assay-report bug 12 or /assay-report feature skill-ingest.

Output

Markdown file saved to /Users/aliz/Documents/Projects Brain/at/assay/Cards/{type}-{id}/{type}-{id}-completion.md.

Vault rule: mkdir -p the card folder, add frontmatter (project: assay, kind: completion, type, id, date, tags) and Part of [[assay]]. Quote the path (it has a space).

The template

## #{id} - {Feature or area} - {Short Title}

Hey! Quick note that this one's {verb}.

**What changed (in plain words):**
{One or two friendly sentences. No jargon. No em dashes.}

**See it live:** {link to the PR, or the CLI command}
**Screenshot/video:** {attached, or link}

If anything looks off or you'd like a tweak, just say the word.

Verb per type: bug → fixed, task → done, feature → shipped.

Tone rules

  • Plain English. Swap declared scope, observed behavior, sandbox, capture, diff, finding, verdict, bundle, content-addressed for everyday phrasing. "Runs the skill and checks it only does what it says" beats "execution-grounded declared-vs-observed diffing."
  • No checklists, no commit hashes, no file paths.
  • Never use the em dash character —.

Don't over-claim (this one matters here)

  • Don't say a skill is "safe" when the verdict is CLEAN — say the observed run matched what it claimed, this time.
  • Don't imply MALICIOUS without saying it was reproduced, captured behavior — never an accusation based on how a skill reads.
  • If the work is a slice of a bigger capability, say it's a first step.

Workflow

  1. Get the context. gh issue view "$ID" or gh pr view "$PR", or write from the session's own work if neither resolves.
  2. Distill into one or two plain sentences.
  3. Pick the "See it live" target: the merged PR, or the exact CLI command (uv run assay check <skill>).
  4. Ask for a screenshot or short video if one isn't on hand.
  5. Write the note and tell the user it's ready.

Example (feature)

## #34 - Scanner - Check a skill before you install it

Hey! Quick note that this one's shipped.

**What changed (in plain words):**
Assay can now run a skill in a locked-down sandbox and watch exactly what it
does, then check that against what it claims to do. If it tries something it
never mentioned, that gets flagged instead of slipping through quietly.

**See it live:** run `uv run assay check <skill>` and watch the verdict
**Screenshot/video:** attached

If anything looks off or you'd like a tweak, just say the word.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.