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/thewolffish/wolffish-app/knowledgenpx skills add thewolffish/wolffish-app --skill knowledgegit clone --depth 1 https://github.com/thewolffish/wolffish-appWrote 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/skills/thewolffish/wolffish-app/knowledge)<a href="https://agentmods.dev/skills/thewolffish/wolffish-app/knowledge"><img src="https://agentmods.dev/badge/skills/thewolffish/wolffish-app/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 | $0.00033 | $0.02567 |
| Opus 5 | $0.00016 | $0.01283 |
| Sonnet 5 | $0.00007 | $0.00513 |
| Haiku 4.5 | $0.00003 | $0.00257 |
Grade A, and why
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 — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge — maintain what you believe
Everything you carry long-term is yours to write: add a new belief, amend one that has gone stale, forget one that is no longer true. These seven tools are the only way to do any of it without waiting for the nightly pass. Nine files, one fixed list — nothing else is reachable, and every write keeps a backup.
The files
| Target | Holds | In every prompt |
|---|---|---|
playbook |
Behavioural lessons: Do / Avoid / User likes / User dislikes / Recipes | yes |
instructions |
Standing custom procedures; outrank the core contract | yes |
soul |
Your character, voice, manner | yes |
user |
The durable profile of who the user is | yes |
projects people preferences technical decisions |
Curated long-term facts | headings only |
knowledge_list prints this live, with sizes and section anchors.
Which file — the boundary that matters
- A lesson about how you should behave →
playbook. ("Avoid: don't send the digest before 07:00.") - A fact about the user or the world → the matching knowledge file. ("People → Sana: moved to Riyadh in June.")
- A standing procedure the user dictates →
instructions. ("Always reply in Arabic on LinkedIn.") - Something about your own manner →
soul.
Putting a behaviour rule in a knowledge file makes it invisible to your habits; putting a fact in the playbook burns the character budget that rides in every prompt. When unsure: behaviour → playbook, world → knowledge.
When to reach for these
The trigger is the user telling you to keep, change, or drop something:
- "Remember I always want invoices as PDF" →
knowledge_addtopreferences. - "Sana's birthday is in March" →
knowledge_addtopeopleunder her section. - "From now on, always CC Omar" →
knowledge_addtoinstructions. - "Stop opening every reply with a summary" →
knowledge_addtoplaybookunderAvoid— orknowledge_forgettheDoentry that taught it. - "That's wrong, she works at X now" →
knowledge_editonpeople. - "You keep believing I use Postgres — I moved to SQLite" →
knowledge_editontechnical. - "Forget what I said about the 6am digest" →
knowledge_forgetonplaybook.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 237 lines · 33 tokens per session scan A 6e66ddbdcccf
knowledge is a skill published in the GitHub repository thewolffish/wolffish-app (5 stars, last pushed yesterday), licensed MIT. It adds 33 tokens to every session and 2,567 once invoked, about $0.0002 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.
Other skills, from other repositories
collaboration-intelligence
Maintain and use a cross-channel collaboration map of rooms, people, agents, agent owners, relationships, expertise, ownership, roster, priority/VIP handling, and recent context. Trigger when a task or message carries room/channel/sender/member metadata; the user asks who is in a room, what it is for, who owns or…
context-drop
Tiny Swift CLI that reads the focused app's text selection via the macOS Accessibility API. Sutando.app's "drop context" action (hotkey configurable via state/hotkeys.json) shells out to this binary to capture what you have highlighted.
context-reconstruct
Re-anchor on the durable record (current-track, live owner thread, pending-questions, relay, buildlog) before acting on anything that depends on earlier context. Read, do not recall.
session-recap
Reconstruct what happened in a past core session — from a high-level summary down to verbatim owner quotes — by reading the raw session transcripts (complete, crash-proof, unbiased), not the curated relay/handoff notes.
agent-computer-use
REQUIRED for any task that involves operating a desktop application — opening apps, clicking buttons, typing into fields, pressing keys, scrolling, dragging, reading what's on screen, moving or resizing windows, or verifying state after an action. Always use the agent-cu CLI commands (open, snapshot, click, type, key…
structured-memory
Use whenever persistent agent memory is read, written, or has grown past its budget - the first time memory is consulted in a session, whenever something worth remembering appears, and whenever MEMORY.md or a topic file exceeds its size budget. Implements a two-tier structure (slim MEMORY.md index linking memory-.md…