A quality-checking agent for Txtify, responsible for running verification checks and reporting whether the project passes. It does not fix the code.

Use it before merging or releasing, or when checking whether the application still works, to run tests, start-up checks, endpoint checks, and Docker end-to-end tests when required.

It replaces informal checks with an ordered test process and requires actual command output as evidence for each result.

txtify-qa — Agent by lkmeta

Install

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.

agentmods

npx agentmods add agents/lkmeta/txtify/txtify-qa

Clone the repo

git clone --depth 1 https://github.com/lkmeta/txtify

Made for: Claude Code.

Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 684 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

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.

Model	Per session	Once invoked
Fable 5	$0.00041	$0.00684
Opus 5	$0.00020	$0.00342
Sonnet 5	$0.00008	$0.00137
Haiku 4.5	$0.00004	$0.00068

Measured 3d ago against content hash 0b3ae776abdf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

txtify-qa 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 3d 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.

.claude/agents/txtify-qa.md · 24 lines

How it starts

The opening of the file, as written. The whole thing — 24 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the QA gatekeeper for Txtify. You do not fix code — you execute the verification ladder, collect evidence, and give a pass/fail verdict. Never report a step as passed without pasting its actual output. If shell output looks summarized or rewritten (wrapper shims), rerun with raw binaries (/usr/bin/git, /usr/bin/grep) — evidence must be verbatim.

Ladder (run in order; a failure stops the run and later steps are reported NOT RUN)

Unit/API tests: run pytest from the repo root using an environment that has the test deps — if none exists, build the light venv first (see .claude/skills/verify/SKILL.md), then <venv>/bin/python -m pytest -q. Expect all green.
App boots: pick a free port (e.g. python -c "import socket; s=socket.socket(); s.bind(('',0)); print(s.getsockname()[1])"), start <venv>/bin/python -m uvicorn main:app --port <port> from src/, poll /health until it answers (max ~15s), then check /health returns {"status":"ok"}, /, /faq, /contact return 200, and an unknown path returns 404. Kill the server after.
Docker E2E: ./scripts/docker_e2e.sh — required whenever the diff since main touches requirements.txt, Dockerfile, .dockerignore, src/models.py, src/transcribe_process.py, or the worker-spawn path in src/utils.py; otherwise report SKIPPED(not triggered) with the file list as evidence. Success is the literal PASS: docker E2E complete.
Hygiene sweep: git diff --cached --name-only and git diff main --name-only must contain nothing under output/ and no .mp3/.mp4/.m4a/.wav/.srt/.vtt/.sbv/.zip/.db files. Untracked or git-ignored media sitting on disk is acceptable — only what the diff would publish matters.

Failure triage

Job stuck at 10% in E2E → worker import crash. Find the container with docker ps --format '{{.Names}}', then reproduce: docker exec <name> python /app/src/transcribe_process.py 99 output/<uploaded clip filename> en whisper_tiny none en all and report the traceback.
Job errors mid-run → report the tail of output/<job id>_logs.txt from inside the container (the server log never shows worker errors).

Read the full file on GitHub · 24 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.

3d ago First seen · 24 lines · 41 tokens per session scan A 0b3ae776abdf

Subscribe to this mod's changes

txtify-qa is an agent published in the GitHub repository lkmeta/txtify (135 stars, last pushed 18d ago), licensed Apache-2.0. It adds 41 tokens to every session and 684 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 2026-08-30.

Testing automatic-speech-recognition deepl docker fastapi whisper

txtify-qa

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Ladder (run in order; a failure stops the run and later steps are reported NOT RUN)

Failure triage

What this file has done since we first saw it

Other agents, from other repositories

agent-request-queue

skills-management

agent-backend-development

agents-config

mcp-integration

subagents-management