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 skills/microsoft/vscode/feedback-learningnpx skills add microsoft/vscode --skill feedback-learninggit clone --depth 1 https://github.com/microsoft/vscodeWhat 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 | $0.00021 | $0.00804 |
| Opus 5 | $0.00010 | $0.00402 |
| Sonnet 5 | $0.00004 | $0.00161 |
| Haiku 4.5 | $0.00002 | $0.00080 |
Grade A, and why
feedback-learning 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feedback learning
Use this skill when a user explicitly corrects an implementation or design approach, rejects a pattern, or asks the agent to learn from feedback, except when the user invokes the literal learn! trigger.
Literal learn! requests are governed exclusively by .github/instructions/learnings.instructions.md and are outside this skill's scope. Do not replace or reinterpret that instruction.
Goal
Preserve reusable knowledge in the smallest authoritative artifact. A correction is not automatically a permanent rule.
Workflow
-
Identify the correction
- What was wrong?
- Why was it wrong?
- What approach did the user prefer?
- Which paths or subsystem does it affect?
-
Search before writing
- Check applicable instructions, skills, design documents, tests, and scoped learning files.
- Update an existing rule instead of adding a duplicate.
-
Classify the feedback
Feedback kind Destination Task-specific preference or one-off adjustment Do not persist Concrete behavior that must not regress Regression test Stable subsystem architecture or product invariant Relevant design/specification document Universal repository rule Applicable instruction file Reusable but not yet mature or proven guidance .github/learnings/<area>.mdTool-driven workflow Relevant skill -
Generalize carefully
- Preserve the principle, not the incident chronology.
- Do not include temporary symbol names, line numbers, or implementation details unless they define the durable contract.
- Do not turn a single rejected implementation into a universal prohibition without broader evidence.
-
Record once
- Design documents and tests are authoritative.
- A learning inbox entry is temporary. When promoted, remove the inbox entry in the same change.
- Never copy the same rule into a skill, instruction, and design document.
- Before adding an inbox entry, compact the target file using the maintenance rules below.
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.
- yesterday First seen · 93 lines · 21 tokens per session scan A 49d47a02f549
feedback-learning is a skill published in the GitHub repository microsoft/vscode (190,061 stars, last pushed yesterday), licensed MIT. It adds 21 tokens to every session and 804 once invoked, about $0.0001 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.
Other skills, from other repositories
cyrene-learn-tutor
Cyrene 在 Learn 模式下的教学流程策略：陪伴用户理解材料、生成练习、检查理解、维护学习进度。.
self-improving-agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge…
blog-post
Writes and structures long-form blog posts, creates tutorial outlines, and optimizes content for SEO with cover image generation. Use when the user asks to write a blog post, article, how-to guide, tutorial, technical writeup, thought leadership piece, or long-form content.
remember
Review the current conversation and capture valuable knowledge — best practices, coding conventions, architecture decisions, workflows, and user feedback — into persistent memory (AGENTS.md) or reusable skills. Use when the user says: (1) remember this, (2) save what we learned, (3) update memory, (4) capture…
pr-finalize
Finalizes any PR for merge by verifying title/description match implementation AND performing code review for best practices. Use when asked to "finalize PR", "check PR description", "review commit message", before merging any PR, or when PR implementation changed during review. Do NOT use for extracting lessons (use…
learn-from-pr
Analyzes a completed PR to extract lessons learned from agent behavior. Use after any PR with agent involvement - whether the agent failed, succeeded slowly, or succeeded quickly. Identifies patterns to reinforce or fix, and generates actionable recommendations for instruction files, skills, and documentation.