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.
git clone --depth 1 https://github.com/deapi-ai/claude-code-skillsWrote 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/commands/deapi-ai/claude-code-skills/transcribe)<a href="https://agentmods.dev/commands/deapi-ai/claude-code-skills/transcribe"><img src="https://agentmods.dev/badge/commands/deapi-ai/claude-code-skills/transcribe/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.
<a href="https://agentmods.dev/commands/deapi-ai/claude-code-skills/transcribe"><img src="https://agentmods.dev/badge/commands/deapi-ai/claude-code-skills/transcribe.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00026 | $0.00562 |
| Opus 5.5 | $0.00010 | $0.00225 |
| Sonnet 5.5 | $0.00005 | $0.00112 |
| Haiku 4.5 | $0.00003 | $0.00056 |
Grade A, and why
transcribe scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -X POST "https://api.deapi.ai/api/v1/client/vid2txt" \ This is a copy
100% identical to transcribe — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
Transcription via deAPI
Transcribe media from URL: $ARGUMENTS
Step 1: Determine endpoint
| URL type | Endpoint |
|---|---|
.mp3, .wav, .m4a, .flac, .ogg |
aud2txt |
YouTube, X, .mp4, .webm, other video |
vid2txt |
Step 2: Send request
For video:
curl -s -X POST "https://api.deapi.ai/api/v1/client/vid2txt" \
-H "Authorization: Bearer $DEAPI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"video_url": "$ARGUMENTS", "include_ts": true, "model": "WhisperLargeV3"}'
For audio:
curl -s -X POST "https://api.deapi.ai/api/v1/client/aud2txt" \
-H "Authorization: Bearer $DEAPI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"audio_url": "$ARGUMENTS", "include_ts": true, "model": "WhisperLargeV3"}'
Step 3: Poll status (feedback loop)
Extract request_id from response, then poll every 10 seconds:
curl -s "https://api.deapi.ai/api/v1/client/request-status/{request_id}" \
-H "Authorization: Bearer $DEAPI_API_KEY"
Status handling:
processing→ wait 10s, poll againdone→ proceed to Step 4failed→ report error message to user, STOP
Step 4: Fetch and present result
When status = "done", fetch content from result_url and present transcription in readable format with timestamps.
Step 5: Offer follow-up
Ask user: "Would you like me to summarize the main points?"
Error handling
| Error | Action |
|---|---|
| 401 Unauthorized | Check if $DEAPI_API_KEY is set correctly |
| 429 Rate Limited | Wait 60s and retry |
| 500 Server Error | Wait 30s and retry once |
| Invalid URL | Ask user to verify the URL format |
| Timeout (>5min) | Inform user, suggest trying shorter content |
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.
- 1mo ago First seen · 67 lines · 26 tokens per session scan A b8c9ba85e4a7
transcribe is a command published in the GitHub repository deapi-ai/claude-code-skills (22 stars, last pushed 3mo ago), licensed MIT. It adds 26 tokens to every session and 562 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). It is 100% identical to transcribe, differing in 0 lines, and is treated as a copy.
Other commands, from other repositories
tray-referencia-api
Referência rápida de todos os endpoints da API da Tray, aceita nome do recurso como filtro.
learn
Capture what the user changed in a draft the skill wrote, so the loop can learn from it. Every rewrite, cut, or addition is evidence; this command turns it into ledger records that /humanise improve mines for rule-change candidates.
fingerprint
(Re)generate the user's voice fingerprint from their corpus. Run after adding profile/sample-.md samples.
code-review
Review code quality applying DRY, SOLID, naming, and formatting standards.
perf-investigate
Drive an iOS performance or memory-leak investigation on using Memory Graph + Time Profiler. Enforce data-before-architecture and stacked PRs.
feature-flag-check
Investigate the feature flag $ARGUMENTS and produce a detailed investigation for removal.