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/jscott3201/aionforge-memory/work-trackingnpx skills add jscott3201/aionforge-memory --skill work-trackinggit clone --depth 1 https://github.com/jscott3201/aionforge-memoryWrote 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/jscott3201/aionforge-memory/work-tracking)<a href="https://agentmods.dev/skills/jscott3201/aionforge-memory/work-tracking"><img src="https://agentmods.dev/badge/skills/jscott3201/aionforge-memory/work-tracking.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 | $0.00075 | $0.00970 |
| Opus 5 | $0.00037 | $0.00485 |
| Sonnet 5 | $0.00015 | $0.00194 |
| Haiku 4.5 | $0.00007 | $0.00097 |
Grade A, and why
work-tracking 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 3d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Work Tracking
Requires an enabled Aionforge Memory MCP server.
Use this skill to make the work durable, not just the facts about it. A work item is a first-class node: persistent, status-tracked, and exempt from the decay and forgetting that apply to memory episodes. Open one as soon as a task, blocker, plan step, or follow-up appears — and keep its status current as the work moves.
Work Item vs Memory — pick the right node
This is the most common mistake, so decide deliberately:
- A task, blocker, TODO, plan step, or follow-up is a work item: use
work_create, thenwork_advance. It persists until you finish or drop it. - A durable fact, decision, validation result, or handoff is a memory
episode: use
capture(see thememory-captureskill). It decays and can be forgotten. - There is no "note" to store directly. Notes are derived by
consolidatefrom episodes — never written by hand. A free-floating "note" is either a fact (capture) or a thing to do (work_create).
Procedure
- Resolve identity once: prefer
AIONFORGE_AGENT_ID; otherwise use the stable agent UUID from the user or project instructions. Mutating tools takeviewer: agent:<uuid>(or an explicitprincipal). - When a task or blocker appears,
work_createit: give a cleartitle, alevel(open vocabulary — e.g. epic, story, task, chapter), and an optionalbody. Nest it withparent_idto build a tree (parent and child must share a namespace). New items start attodo. - Advance status as the work moves with
work_advance:todo → in_progress → blocked → done, ordroppedwhen abandoned. Passexpected_fromfor a guarded compare-and-set when you need to avoid clobbering concurrent progress. This is the only audited work op, so it is the system of record for what happened. - Classify with
work_linkwhen a controlled-vocabulary tag helps (slug, optionaldisplay); it is idempotent and mints the tag on first use. - Read the backlog back with
work_query(filter bywork_statusand/orlevel) and a subtree withwork_tree(aroot_idplus adepth). Recall the state before assuming it. Assert the teams you belong to on these reads (e.g.teams: ["aionforge-memory-team"]): read authorization is per-call, so a team's work items are out of scope unless you assert that team in the same call — and a by-idread_memoryof a team work item likewise requires the team asserted in that call (parity with search). Never assert a team you are not a member of. - Default to a private item (omit
target_namespace). Use a sharedtarget_namespace(e.g.team:project-alpha) only when the host or user authorizes that scope.
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.
- 3d ago First seen · 71 lines · 75 tokens per session scan A de20548bb7e8
work-tracking is a skill published in the GitHub repository jscott3201/aionforge-memory (10 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 75 tokens to every session and 970 once invoked, about $0.0004 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
zettelforge
ZettelForge v2.0.0 — Production CTI agentic memory system. Hybrid TypeDB (STIX 2.1 ontology) + LanceDB (vector search). Zero external AI dependencies: fastembed for embeddings, llama-cpp-python for LLM. 75% accuracy on CTI queries, 18% on LOCOMO. Use when agents need persistent memory, threat intel retrieval, entity…
matryca-github
Apply Matryca maintainer standards for GitHub issues, pull requests, branches, reviews, merges, milestones, tags, releases, and remote comments. Use before any GitHub-facing action or artifact.
matryca-changelog
Decide whether a completed Matryca change belongs in CHANGELOG.md and add one concise Unreleased entry when required. Use before concluding runtime, security, architecture, integration, performance, operator, or public-contract changes.
matryca-clean-architecture
Preserve Matryca Clean Architecture dependency direction, thin surfaces, domain ownership, typed boundaries, config parsing, and scoped refactors. Use for module moves, shared abstractions, layer imports, configuration architecture, or structural Python changes.
matryca-release
Prepare, cut, verify, tag, or publish a Matryca semantic-version release. Use for version bumps, changelog finalization, release notes, build metadata, tags, GitHub Releases, or PyPI publication.
matryca-env
Keep Matryca environment-variable code, defaults, UI mappings, .env.example, coverage tests, and changelog synchronized. Use when adding, renaming, deprecating, exposing, or changing any environment variable or default.