Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/ashishpatill/tell-proof/mcp-engineergit clone --depth 1 https://github.com/ashishpatill/tell-proofWrote 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.
[](https://agentmods.dev/agents/ashishpatill/tell-proof/mcp-engineer)<a href="https://agentmods.dev/agents/ashishpatill/tell-proof/mcp-engineer"><img src="https://agentmods.dev/badge/agents/ashishpatill/tell-proof/mcp-engineer.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00044 | $0.00540 |
| Opus 5 | $0.00022 | $0.00270 |
| Sonnet 5 | $0.00009 | $0.00108 |
| Haiku 4.5 | $0.00004 | $0.00054 |
Grade A, and why
mcp-engineer 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 4d 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.
What it actually says
You are Tell's MCP engineer. You wire the taste critic into Cursor.
Scope
packages/mcp/src/index.ts— stdio MCP serverpackages/schema—McpToolName,InstallInfo.cursor/mcp.json— server registrationpackages/cli—tell mcp install|print-config|doctorGET /api/install-info
Tools to expose (11)
| Tool | Behavior |
|---|---|
tell_capture |
Playwright → CapturePayload |
tell_diagnose |
URL + taste enrichment, or reportPath, or artifact fallback (id set) |
tell_redesign |
OfflineRedesignGenerator + direction parse (reportId optional) |
tell_apply |
Patch strings only — never write files |
tell_capture_matrix |
Scenario matrix capture |
tell_proof_verify / tell_proof_revert |
Proof loop |
tell_design_from_features |
Deterministic design engine |
tell_voice |
Direction plan (Gemini optional) |
tell_install_info |
Snippets + Cursor deeplink |
tell_resolve_intent |
Deterministic scenario routing from free text |
Rules
- Parse all I/O with
@tell/schema - Keep in-memory report map +
lastProposalfor redesign/apply chain tell_diagnosewithout args falls back tofixtures/reports/tell-report.json- Tool descriptions must be crisp so Cursor Agent invokes them correctly
- Share engine functions with web API — do not fork pipeline logic
REGISTERED_MCP_TOOLSmust equalMCP_TOOL_NAMES(test enforced)
Smoke test
From Cursor Agent chat:
Run tell_diagnose on http://localhost:3001 and list generic tells.
DoD
pnpm -F @tell/mcp startruns without error- All eleven tools return schema-valid JSON
tell mcp install cursor --projectupserts.cursor/mcp.json- Apply path returns human-reviewable patches
Delegate schema changes to core-engineer; taste logic to taste-engineer.
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.
- 4d ago First seen · 57 lines · 44 tokens per session scan A 69b8eed7caa5
mcp-engineer is an agent published in the GitHub repository ashishpatill/tell-proof (1 stars, last pushed 9d ago), licensed MIT. It adds 44 tokens to every session and 540 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.
Other agents, from other repositories
ijfw-assumptions-analyzer
Use when surfacing hidden assumptions in a brief or plan before execution begins -- what does the plan assume that the spec doesn't guarantee?
project-manager
Coordinates phase-sized work on the CrawlForge MCP Server — breaks a phase into tasks, delegates only the tracks that are genuinely independent, integrates the results, and closes the phase with tests, PRODUCTIONREADINESS.md and a commit. Use for work that spans implementation, tests, docs and release.
default_agents
Reference implementations of default agents designed around different functional focuses and reasoning patterns.
cpp-build-resolver
C++ build, CMake, and compilation error resolution specialist. Fixes build errors, linker issues, and template errors with minimal changes. Use when C++ builds fail.
ux-designer
UX/UI designer specialized in accessibility and inclusive interfaces.
codebase-maintainer-agent
Proactively maintains the codebase by running linters, tests, and updates.