distill

distill is a skill for Claude Code, Codex from tianhanz/ears. It costs 133 tokens per session (696 once invoked), scanned A, original, MIT.

A tool for finding recurring lessons in trace files, which are records of errors, surprises, choices, and results. It turns patterns found across multiple traces into reusable knowledge files.

In plain words
What is it for?
Use it to scan traces, identify repeated problems or successful settings, assess their strength, and write structured knowledge entries.
Why use it?
It helps preserve useful discoveries instead of leaving them scattered across individual debugging records. It also records how often and how reliably each pattern appears.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions CLAUDE.md.

Good fit Use it to scan traces, identify repeated problems or successful settings, assess their strength, and write structured knowledge entries.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/tianhanz/ears/distill
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 tianhanz/ears --skill distill
Clone the repo
git clone --depth 1 https://github.com/tianhanz/ears

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 distill

README.md
[![agentmods](https://agentmods.dev/badge/skills/tianhanz/ears/distill.svg)](https://agentmods.dev/skills/tianhanz/ears/distill)
Your own site
<a href="https://agentmods.dev/skills/tianhanz/ears/distill"><img src="https://agentmods.dev/badge/skills/tianhanz/ears/distill.svg" alt="Measured on agentmods" height="20"></a>
Per session 133 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 696 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 $0.00133 $0.00696
Opus 5 $0.00067 $0.00348
Sonnet 5 $0.00027 $0.00139
Haiku 4.5 $0.00013 $0.00070

Measured 7d ago against content hash f77c66cdf30d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

distill 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 7d 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/distill/SKILL.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/distill — Pattern Extraction from Traces

Extract patterns from trace.md files into reusable knowledge files. Turns scattered experience into distilled, actionable knowledge.

Trigger

User mentions: "distill", "extract patterns", "what have we learned", "summarize traces", "knowledge distillation". Also use after completing a batch of reproductions to extract emergent knowledge.

Workflow

Step 1 — Gather traces

Search for trace.md files across the project:

find . -name 'trace.md' -not -path './.git/*' | sort

Read each trace file. Look for:

  • Errors and their root causes
  • Surprising results or unexpected behaviors
  • Parameter choices that worked (or didn't)
  • Workflow decisions and their outcomes

Step 2 — Identify patterns

A pattern is a recurring observation that appears in 2+ traces. Tag each pattern:

[N=<count>, <weak|moderate|strong>]
  • weak: 2 occurrences, may be coincidence
  • moderate: 3-4 occurrences with consistent mechanism
  • strong: 5+ occurrences, well-understood mechanism

Quantitative patterns with specific thresholds beat qualitative summaries. Example:

  • Good: "Grid resolution of 15+ points per characteristic length scale is needed for 1% accuracy [N=7, strong]"
  • Bad: "Use fine grids [N=7, strong]"

Step 3 — Write concept file

Create knowledge/<concept>.md with:

# <Concept Name>

## Established Patterns [N >= 3]
- Pattern 1 [N=5, strong] — Source: trace1, trace2, ...
- Pattern 2 [N=3, moderate] — Source: ...

## Emerging Patterns [N = 2]
- Pattern 3 [N=2, weak] — Source: ...

## Open Questions
- Question that traces surface but don't answer

## Action Items
- [ ] Concrete steps to validate or apply these patterns

Step 4 — Check for promotion

If any pattern reaches [N >= 5, strong], it is a candidate for promotion to CLAUDE.md or other project-level documentation. Flag these to the user — only the project owner decides what goes into the foreground.

Anti-Patterns

  • Don't distill from a single trace. Wait for patterns to emerge from 2+ traces.
  • Don't write platitudes ("be careful with parameters"). Be specific and quantitative.
  • Don't duplicate existing knowledge. Check concept files before writing.
  • Don't promote prematurely. A pattern at N=2 is weak — it may be coincidence.

Read the full file on GitHub · 76 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. 7d ago First seen · 76 lines · 133 tokens per session scan A f77c66cdf30d

Subscribe to this mod's changes

distill is a skill published in the GitHub repository tianhanz/ears (5 stars, last pushed 4mo ago), licensed MIT. It adds 133 tokens to every session and 696 once invoked, about $0.0007 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 2026-08-31.

Related

Other skills, from other repositories

media-ingest

Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.

garrytan/gbrain · 52 tokens

mem0-oss-to-platform

Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…

mem0ai/mem0 · 273 tokens

Cortex

Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…

danielmiessler/LifeOS · 196 tokens

memory

Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.

opensquilla/opensquilla · 44 tokens

ha-data-stores

Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…

shiwenwen/hope-agent · 115 tokens

establishing-project-context

Use when the user asks to establish shared project language, or project work exposes a conflicting, renamed, or deprecated domain term that needs active semantic modeling. Routine small tasks stay on the fast path.

GanyuanRan/Aegis · 45 tokens