feedback-capture

feedback-capture is a skill for Claude Code, Codex from tigu77/tiguclaw. It costs 130 tokens per session (1,692 once invoked), scanned A, original, Apache-2.0.

A feedback-memory skill that turns clear user corrections or confirmations about an answer into a short reusable instruction. It ignores ordinary thanks and records ambiguous reactions without guessing their meaning.

In plain words
What is it for?
Use it when a user says what to stop doing, what to keep doing, or clearly confirms a chosen style; it can ask for clarification when the feedback is unclear.
Why use it?
It helps preserve useful preferences across later conversations without treating every positive or negative remark as a rule.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it when a user says what to stop doing, what to keep doing, or clearly confirms a chosen style; it can ask for clarification when the feedback is unclear.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/tigu77/tiguclaw/feedback-capture
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.

Any agent
npx skills add tigu77/tiguclaw --skill feedback-capture
Clone the repo
git clone --depth 1 https://github.com/tigu77/tiguclaw

Made for: Claude Code, Codex.

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 feedback-capture

README.md
[![agentmods](https://agentmods.dev/badge/skills/tigu77/tiguclaw/feedback-capture/github.svg)](https://agentmods.dev/skills/tigu77/tiguclaw/feedback-capture)
Your own site
<a href="https://agentmods.dev/skills/tigu77/tiguclaw/feedback-capture"><img src="https://agentmods.dev/badge/skills/tigu77/tiguclaw/feedback-capture/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 feedback-capture

Your own site · 80×15
<a href="https://agentmods.dev/skills/tigu77/tiguclaw/feedback-capture"><img src="https://agentmods.dev/badge/skills/tigu77/tiguclaw/feedback-capture.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 130 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,692 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 $0.00130 $0.01692
Opus 5.5 $0.00052 $0.00677
Sonnet 5.5 $0.00026 $0.00338
Haiku 4.5 $0.00013 $0.00169

Measured 2d ago against content hash 3f0219fe897a, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

feedback-capture 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 2d 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.

skills/feedback-capture/SKILL.md · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

반응으로 배우기

사용자의 «좋다»·«이렇게 하지 마» 는 다음 대화부터 지켜져야 의미가 있다. 그 자리에서 한 줄 규칙으로 만들고 사용자 앞에서 매 턴 실리는 확정 지침(SELF_GROWTH.md)에 올린다 — 분명하면 올리고 밝히고, 애매하면 묻고 올린다.

1. 무엇에 대한 반응인가

반응이 어느 응답의 어느 부분을 두고 한 말인지 먼저 짚는다(형식·길이·말투·판단·범위·순서 등). 못 짚겠으면 추측해 규칙을 만들지 말고 짧게 묻는다: «방금 답에서 어느 부분이 과했을까요?»

2. 세기를 가른다

반응 할 일
규칙을 직접 말해 줌 — «앞으로는 ~해», «반말로 해», «이름은 X 야» 처럼 사용자가 규칙 자체를 일반형으로 말함 이 스킬이 아니다 — 헌법대로 묻지 말고 AGENT.md 에 바로 반영한다(사용자가 직접 적어 준 지침)
분명한 교정 — 방금 응답을 두고 «이렇게 하지 마», «그게 아니야», «너무 과해», 같은 지적의 반복 3으로 — 무엇이 문제였는지 내가 정리해야 규칙이 된다
분명한 확인 — 내가 일부러 고른 방식을 «좋다·계속 그렇게» 3으로 — «유지할 것» 규칙
애매한 반응 — «음…», «별로네» 처럼 무엇이 문제인지 안 드러남 규칙을 만들지 않는다. add_memory({type:'feedback', name:'feedback-obs-<주제>-<날짜>'}) 로 증거만 남긴다(사용자 원 발언과 내 추정을 구분)
평범한 맞장구·감사 — «고마워», «응», 평소 하던 걸 칭찬 아무것도 남기지 않는다

★칭찬을 버리지 마라 — 교정만 쌓이면 비서가 억제 한쪽으로만 기운다. 하지만 매번 묻지도 마라 — «좋네» 한마디마다 «기억할까요?» 를 붙이면 그게 새 불만이 된다.

3. 한 줄 규칙으로 정리한다 — 묻는 건 판단이 애매할 때만

  • 상황을 앞에 둔 한 줄: «변경 기록을 쓸 때: 예시·수치 인용을 넣지 않는다», «선택지를 보일 때: 하나를 추천하고 이유를 붙인다».
  • 사용자가 말한 언어로 적는다 — 번역하지 않는다(2026-10-14 정태님. 다국어 검증에서 영어·일본어 교정이 한국어로 번역돼 적히고 스페인어는 그대로 적혀 한 파일에 섞였다).
  • 일반화는 사용자가 말한 범위까지만. 한 번의 지적을 모든 상황으로 넓히지 마라.
  • 범위도 말한다 — 이 설치본 전체인가, 이 프로젝트만인가.
  • ★판단이 분명하면 묻지 말고 올린다 (2026-10-14 정태님: «애매한 것들은 당연히 확인이 필요하고 확실한 건 그냥 하면 되는 거고 — 애매하다는 건 판단이 애매한 부분»). 분명함 = 무엇이 문제였는지와 어디까지 적용할지가 반응에 드러남. 올린 뒤 한 줄로 밝힌다: «앞으로 ~할 때는 ~하겠습니다(기억해 뒀습니다 — 아니면 말씀해 주세요).» «기억할까요?» 를 매번 붙이면 그게 새 불만이 된다(회사돌쇠 10-14: 자료 전달 방식 교정 뒤 «넣을까요?»).
  • 판단이 애매하면 그때만 묻는다: 문제가 된 부분이 둘 이상으로 읽히거나, 한 번의 지적을 어디까지 넓힐지 갈릴 때. «앞으로 ~할 때는 ~하겠습니다. 이렇게 기억할까요?» (선택지 버튼을 써도 된다)
  • ★애매한데 추측으로 올리지 않는다 — 잘못 일반화한 규칙이 매 턴 실리면 고치려던 것보다 나쁘다.

4. 올린다

  • 이 설치본 전체 → confirm_feedback({ key, text }). key 는 상황 이름(소문자·숫자·-·_).
  • 한 프로젝트만 → 여기 말고 그 프로젝트 문서(PROJECT.md 등)에 적는다. 확정 지침은 모든 대화에 실려 다른 프로젝트를 흐린다.
  • 정체성·말투 자체(이름·호칭·말투) → 그 내용을 AGENT.md 에 반영한다(인격은 거기 한 곳에 둔다).
  • 결과 줄에 나오는 개수·바이트를 본다. 상한에 가까우면 다음 단계(합치기)를 먼저 한다.

Read the full file on GitHub · 63 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.

  1. 2d ago First seen · 63 lines · 130 tokens per session scan A 3f0219fe897a

Subscribe to this mod's changes

feedback-capture is a skill published in the GitHub repository tigu77/tiguclaw (9 stars, last pushed yesterday), licensed Apache-2.0. It adds 130 tokens to every session and 1,692 once invoked, about $0.0005 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 2026-10-15.

Related

Other skills, from other repositories

commonly

You are a member of a Commonly workspace — a shared space where humans and AI agents from any origin collaborate in pods (chat rooms with memory). Use this whenever you are connected to Commonly via the commonly MCP tools: to read what's happening, post, remember things across sessions, react, DM other agents, and…

Team-Commonly/commonly · 0 tokens

dbrain-processor

Personal assistant for processing daily voice/text entries from Telegram. Classifies content, saves thoughts to Obsidian with wiki-links, generates HTML reports. Integrates Your Business context (clients, projects, CRM). Triggers on /process command or daily 21:00 cron.

smixs/agent-second-brain · 60 tokens

autograph

Schema-as-code enforcement for any Obsidian vault. Zero hardcoded domains. Use when creating vault cards, checking vault health, running schema compliance, deduplicating entities, generating MOC indexes, running decay cycles, bootstrapping a vault, fixing wikilinks, finding orphans or backlinks, extracting entities…

smixs/agent-second-brain · 89 tokens

claude-md-improver

Audit and improve CLAUDE.md files in repositories. Use when asked to check, audit, update, improve, or fix CLAUDE.md files, or when someone mentions project memory optimization.

bearlike/Assistant · 46 tokens

remember

Curates the current conversation into ByteRover's context tree for future retrieval. Use after important decisions, project discussions, or technical deep-dives.

ahkedia/lyra-ai · 32 tokens

using-axon

Use when operating an installed Axon through its MCP tool or CLI to search, index sources, query the corpus, extract data, manage jobs or watches, and troubleshoot the RAG stack. For first-time setup or deployment repair, use install-axon.

dinglebear-ai/axon · 55 tokens