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/IgorGanapolsky/ThumbGateWrote 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/igorganapolsky/thumbgate/capture-feedback)<a href="https://agentmods.dev/commands/igorganapolsky/thumbgate/capture-feedback"><img src="https://agentmods.dev/badge/commands/igorganapolsky/thumbgate/capture-feedback/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/igorganapolsky/thumbgate/capture-feedback"><img src="https://agentmods.dev/badge/commands/igorganapolsky/thumbgate/capture-feedback.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 | $0.00010 | $0.00281 |
| Opus 5.5 | $0.00004 | $0.00112 |
| Sonnet 5.5 | $0.00002 | $0.00056 |
| Haiku 4.5 | $0.00001 | $0.00028 |
Grade A, and why
capture-feedback 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 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.
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
Capture Feedback
Quickly capture structured feedback about the current task or action.
Usage
Invoke this command to record a feedback signal with context and tags. If the user only gives a quick thumbs_down, wrong, or correct, the command should still call capture_feedback and pass recent context through chatHistory.
Steps
- Specify signal:
thumbs_uporthumbs_down. - If the user already gave a clear explanation, pass it as
context. - If the explanation is vague or omitted, gather up to 8 prior recorded entries plus the failed tool call into
chatHistory. - Add tags for categorization.
- If this is a later clarification for an existing feedback event, include
relatedFeedbackId. - Call the
capture_feedbackMCP tool and show the proposed lesson or corrective rule back to the user.
Example
/capture-feedback thumbs_down "Tests broke after migration" database,migration
/capture-feedback thumbs_down
In the second example, the command should rely on history-aware distillation instead of refusing the signal outright. If the user adds more detail later, reuse relatedFeedbackId so the linked 60-second follow-up session updates the same record.
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 · 34 lines · 10 tokens per session scan A 9e61ba3ac804
capture-feedback is a command published in the GitHub repository IgorGanapolsky/ThumbGate (29 stars, last pushed yesterday), licensed MIT. It adds 10 tokens to every session and 281 once invoked, about $0.0000 per session on Opus 5.5. 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.
Other commands, from other repositories
red-team
Red-team your safety scanning by generating adversarial variants of a prompt injection attempt.
check-safety
Generate a detailed safety risk report for the current conversation or a specific file.
check-pii
Check text or a file for personally identifiable information and show the redacted version.
scan
Scan the provided text for safety issues using all Sentinel AI scanners.
start
A Korean-language start command for beginning a repair loop after refining a goal, making a backup, giving guidance, and confirming the plan.
checkup
A Korean-language command for scheduled checks that only reports what it finds. It examines previously saved goals but never changes or repairs them.