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 commands/saubakirov/trace-first-starter/tfw-knowledgegit clone --depth 1 https://github.com/saubakirov/trace-first-starterWrote 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/saubakirov/trace-first-starter/tfw-knowledge)<a href="https://agentmods.dev/commands/saubakirov/trace-first-starter/tfw-knowledge"><img src="https://agentmods.dev/badge/commands/saubakirov/trace-first-starter/tfw-knowledge.svg" alt="Measured on agentmods" 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 | $0.00011 | $0.01203 |
| Opus 5 | $0.00005 | $0.00602 |
| Sonnet 5 | $0.00002 | $0.00241 |
| Haiku 4.5 | $0.00001 | $0.00120 |
Grade A, and why
tfw-knowledge 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TFW Knowledge — Knowledge Consolidation Workflow
Role: Coordinator Output: Updated
knowledge/topic files,KNOWLEDGE.md§4 (index),knowledge_state.yamlTrigger: Manual (/tfw-knowledge) or gate in plan.md Step 2 Duration: 5-20 minutes
🔒 ROLE LOCK: COORDINATOR Permitted: reading all project artifacts, writing to KNOWLEDGE.md,
knowledge/topic files,.tfw/knowledge_state.yaml. Forbidden: writing code, modifying RF/REVIEW/RES/HL/TS files (except addingfact-candidates: processedmarker).
Prerequisites
- Read
.tfw/knowledge_state.yaml— current state, last consolidation seq - Read
KNOWLEDGE.md— current knowledge index - Read all topic files in
knowledge/— current verified facts - Read
.tfw/project_config.yaml→tfw.knowledgesection for limits
Phase 1: Orient
- Note
last_consolidation_seq→ compute task range to scan - List topic files with fact counts
- Read conventions.md §10.1 → list all canonical categories. Note which have topic files and which don't
- Present orientation summary to user (include category coverage gaps)
Phase 2: Gather
⚠️ Knowledge ≠ technical documentation. Knowledge is what would be UNKNOWN without the human saying it: vision, priorities, emotions, business context, architectural philosophy, process corrections. Technical implementation details (tools config, API constraints, build errors) belong in tfw-docs.
YES: "primary output = knowledge graph, not docs site", "close as MVP, don't stretch phases" NO: "MkDocs docs_dir cannot be project root", "use directory URLs not .md"
- Scan artifacts for tasks since
last_consolidation_seq:- HL §11 (Strategic Session Insights) — coordinator's captured signals from planning sessions
- RF §8 (Execution Session Insights) — executor's captured signals from live testing (if exists)
- RF, REVIEW, RES §Fact Candidates — standard FC sections
- Review conversation history for the current session — extract facts from user messages not captured in artifacts
- Category coverage check: do any candidates belong to categories without topic files? Don't force facts into existing categories — create new topic file when justified. Ref: conventions.md §10.1
- Present gathered candidates to user:
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.
- 5d ago First seen · 102 lines · 11 tokens per session scan A aa4fb4720447
tfw-knowledge is a command published in the GitHub repository saubakirov/trace-first-starter (29 stars, last pushed 2d ago), licensed MIT. It adds 11 tokens to every session and 1,203 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 commands, from other repositories
om-intake
Process all unread meeting notes in work/meetings/ — reads each file, classifies content, routes to the right vault notes, then clears the inbox.
om-dump
Freeform capture mode. Dump anything — conversations, decisions, incidents, wins, thoughts — and I'll route it all to the right notes with proper templates, frontmatter, and wikilinks.
adopt
Adopt an existing directory as the SuperBrain vault (marks it, indexes existing notes).
capture
Quickly save something as a single clean note in the right place — a link, a thought, an article, a snippet.
broadcast
Run broadcast on a distilled Raw — update related existing pages conversationally.
reflect
Reflect on recent work — log lessons to the journal and propose distilled rule updates for review.