agentleFS
krishhggkrishhgg/t3term/.claude/skills/verify-this/SKILL.md

verify-this

Skill11 starsChanged 27 days agoMITView on GitHub

Prove or disprove one claim about t3term with fresh evidence: state it so a run can disprove it, capture a baseline and a treatment the same way, compare the raw artifacts and return VERIFIED, NOT VERIFIED or INCONCLUSIVE. Use for \"verify this\", \"did this fix it\", \"prove it works\" or \"show me the evidence\" about one behavior. For a whole pull request, use verify-pr.

11 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use control-cli, control-ui and 7 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ElyzeSolutionsSkill

    control-cli

    Build or adapt a local harness to drive, inspect, and profile an interactive CLI or TUI without external services. Use for CLI UX checks, startup regressions, memory leaks, hangs, prompt flows, or terminal demos.

    Also in 4 projects that use this

  • ElyzeSolutionsSkill

    control-ui

    Build or adapt a local browser/CDP harness to drive and inspect a web, IDE, or Electron UI. Use for local UI verification, screenshots, accessibility snapshots, perf profiles, visual diffs, or reproducing UI bugs.

    Also in 4 projects that use this

  • jaydubya818Skill

    fix-merge-conflicts

    Use when a branch needs conflicts resolved and the merged result verified.

    Also in 3 projects that use this

  • jaydubya818Skill

    automate-me

    Use when the user wants working preferences captured in a reusable mode skill.

    Also in 4 projects that use this

  • MicrosoftSkill

    bro

    Re-explain the previous assistant message in a much simpler way, for when the reply made you go 'bro what'. Use /bro to get a simpler version of the last answer in Plain Language + Chicago Manual of Style, with diagrams where they help.

    Also in 4 projects that use this

  • termdockSkill

    recall

    Retrieves cross-session memories about past decisions, lessons, and patterns. Use when user asks about prior work, past decisions, or mentions a feature. Triggers include check memory, what did we decide, how did we solve.

    Also in 4 projects that use this

  • +2 more from jaydubya818

The file

We split this skill into the rules it gives an agent

3 rules. 1 is common, 2 are distinctive, and 0 are very specific to this project.

  • 1

    Common

    Most of the 11 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Don't use it for claims nothing can measure, like "the code is cleaner".

    line 12
  2. Distinctive:

    INCONCLUSIVE: No valid baseline, a noisy signal, a failed run, or a difference between the two setups that spoils the comparison.

    line 32
  3. Common:

    Don't soften a negative result.

    line 34

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “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

Verify this

Verification isn't a recap. It proves or disproves one specific claim with evidence someone else can rerun.

Adapted from cursor-team-kit's verify-this skill (github.com/cursor/plugins, MIT) for this repo.

Don't use it for claims nothing can measure, like "the code is cleaner". Ask for a claim a run can check first.

Steps

  1. State the claim so a run can disprove it. The condition, what you observe and the threshold. "After t3v reconnect, the sidebar's feed dot is green again within 5 s and the open thread's draft is unchanged."
  2. Pick the smallest surface that can disprove it.
    • The TUI: a verify-t3term drive, with captures.
    • The CLI: the command run against a t3v fake server (t3v env prints the line), with stdout, stderr and the exit code.
    • The RPC client: a test in tests/fake_server.rs.
    • One function's behavior: a focused cargo test.
    • CPU, memory or energy: benchmarks/, vetted with the benchmark-checklist skill.
  3. Capture a baseline from the old state: the merge base, the parent commit or the build that shows the bug. t3v build <rev> builds any commit without touching your checkout.
  4. Capture the treatment from the changed state with the same command, scenario, size, color depth and steps. When both runs are live at once, keep each one's path from t3v start --json and pass it with --run or T3V_RUN to every command, because /tmp/t3v-current names only the newest run on the machine.
  5. Compare the raw artifacts: screen captures (diff the .txt files), rendered PNGs, t3v status --json counters, CLI output and exit codes, test output.
  6. Return exactly one verdict.

Verdicts

  • VERIFIED. Baseline and treatment differ the way the claim predicts, by its threshold, with nothing else that explains it.
  • NOT VERIFIED. The behavior didn't change, changed the wrong way or missed the threshold.
  • INCONCLUSIVE. No valid baseline, a noisy signal, a failed run, or a difference between the two setups that spoils the comparison.

Don't soften a negative result. A clear NOT VERIFIED is useful.

Where artifacts go

/tmp/verify-this/<claim-slug>/ with claim.md, baseline/, treatment/, diff/ and verdict.md. t3v start --evidence points a run's captures there. Keep them out of the repository.

Reply

VERIFIED | NOT VERIFIED | INCONCLUSIVE
Claim: <the claim as a run can disprove it>

Evidence:
<artifact>: baseline=<...>, treatment=<...>, difference=<...>, threshold=<...>

Reasoning:
<one short paragraph naming the evidence and anything that could confound it>

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.