tell-proof: Skill for Cursor

.cursor/skills/agency-run-learn/SKILL.md

agency-run-learn is a skill for Cursor from ashishpatill/tell-proof. It costs 36 tokens per session (427 once invoked), scanned A, original, MIT.

An automatic learning step for developers who run the agency process. It updates shared design memory and lessons from development research and completed runs.

In plain words
What is it for?
Use it with agency runs, pass-marking, or a separate re-learning command when maintaining the shared design engine.
Why use it?
It helps future agency runs use knowledge from earlier work without treating ordinary end-user sessions as training data.

Skill for Cursor

Written for Cursor: installed under .cursor/.

This is ashishpatill/tell-proof's own configuration. It tells Cursor how to work on tell-proof itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything tell-proof configures →

Reuse

Borrowing it

Nothing to install: this file belongs to ashishpatill/tell-proof. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/ashishpatill/tell-proof/master/.cursor/skills/agency-run-learn/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/ashishpatill/tell-proof

Made for: Cursor.

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 agency-run-learn

README.md
[![agentmods](https://agentmods.dev/badge/skills/ashishpatill/tell-proof/agency-run-learn/github.svg)](https://agentmods.dev/skills/ashishpatill/tell-proof/agency-run-learn)
Your own site
<a href="https://agentmods.dev/skills/ashishpatill/tell-proof/agency-run-learn"><img src="https://agentmods.dev/badge/skills/ashishpatill/tell-proof/agency-run-learn/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 agency-run-learn

Your own site · 80×15
<a href="https://agentmods.dev/skills/ashishpatill/tell-proof/agency-run-learn"><img src="https://agentmods.dev/badge/skills/ashishpatill/tell-proof/agency-run-learn.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 427 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 $0.00036 $0.00427
Opus 5 $0.00018 $0.00214
Sonnet 5 $0.00007 $0.00085
Haiku 4.5 $0.00004 $0.00043

Measured 10d ago against content hash d3359d84f5bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

agency-run-learn 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 10d 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.

.cursor/skills/agency-run-learn/SKILL.md · 42 lines

What it actually says

agency-run-learn

Developer / maintainer loop. Each agency:run on a workstation with a design-data pointer ends with learn → agency-engine-memory.json + LEARNINGS.md (+ write-back).

Not for Ashish's browser. His preferences use tell-user-session-learn.

Loop Skill Where
Dev corpus + pipeline agency-run-learn research/*, design-data checkout
User sessions tell-user-session-learn localStorage tell:user-design-profile

Automatic on pnpm agency:run and --mark-pass 4-ship. Opt out: AGENCY_SKIP_LEARN=1.
Dev corpus itself requires research/design-data.local.json or TELL_DESIGN_DATA + TELL_DEV_CORPUS=1.

Hard rules

  1. Learn is automatic on agency runs — do not schedule a separate agent step.
  2. Never enable design-data pull/write-back on public demos (VERCEL / TELL_PUBLIC_DEMO).
  3. Do not weaken gates to make a run green.
  4. No third-party hosts in Tell commits.
  5. Do not store end-user profiles in the design-data repo.

Commands

pnpm agency:run -- --query "<requirement>" --fresh   # learn automatic
pnpm agency:learn -- --run-id <id>                   # re-learn only
AGENCY_SKIP_LEARN=1 pnpm agency:run -- --query "…"   # dry smoke
  • research/design-data.README.md — developer corpus wiring
  • tell-user-session-learn — per-user product learning
  • agency-quality-site — phased craft + autonomous run
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. 10d ago First seen · 42 lines · 36 tokens per session scan A d3359d84f5bc

Subscribe to this mod's changes

agency-run-learn is a skill published in the GitHub repository ashishpatill/tell-proof (1 stars, last pushed 15d ago), licensed MIT. It adds 36 tokens to every session and 427 once invoked, about $0.0002 per session on Opus 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 2026-08-31.