Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/andreaswasita/copilot-agents-dojonpx agentmods add skills/andreaswasita/copilot-agents-dojo/durable-workWrote 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/andreaswasita/copilot-agents-dojo/durable-work)<a href="https://agentmods.dev/skills/andreaswasita/copilot-agents-dojo/durable-work"><img src="https://agentmods.dev/badge/skills/andreaswasita/copilot-agents-dojo/durable-work/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/andreaswasita/copilot-agents-dojo/durable-work"><img src="https://agentmods.dev/badge/skills/andreaswasita/copilot-agents-dojo/durable-work.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Memory Poisoning · line 35 Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.Fix: Implement context-window management that detects and rejects padding or stuffing attempts. Prioritize system instructions over user-injected content.
- medium Memory Poisoning · line 65 Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.Fix: Implement context-window management that detects and rejects padding or stuffing attempts. Prioritize system instructions over user-injected content.
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 5.1 | $0.00017 | $0.01327 |
| Opus 5 | $0.00009 | $0.00664 |
| Sonnet 5 | $0.00003 | $0.00265 |
| Haiku 4.5 | $0.00002 | $0.00133 |
Grade A, and why
durable-work 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 9d 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Durable Work Skill
Routes work that must survive a single Copilot turn to tasks/board/ instead of the task tool. Sub-agents launched via task are killed if the parent is cancelled (rate-limit, user navigation, OS sleep) — anything you need across sessions belongs on the durable board. The complementary skill is skills/subagent-strategy, which handles in-turn delegation.
When to Use
- Work estimated > 1 turn (e.g. a multi-step deployment, a long build, an iterative review).
- Work that multiple agents or a human will hand off across sessions.
- Work whose state must be queryable later (
scripts/board.sh status). - Work whose loss would block the user (e.g. partial migrations, in-flight refactors).
- NOT for: one-shot research, fast file reads, parallel grep — those are sub-agent territory.
Prerequisites
tasks/board/exists (created byscripts/init.shor the dojo template).scripts/board.shavailable on PATH or asbash scripts/board.sh.- Familiarity with
.dojo/delegation.yaml→escalate_to_board_iftriggers.
How to Run
1. Decide: does this work need to survive parent cancellation? If yes, board.
2. Create the task file: bash scripts/board.sh new "<title>".
3. Fill Context, Plan, Verification in the new tasks/board/NNN-slug.md.
4. Work through Plan checkboxes; update `status:` frontmatter as you go.
5. On completion: set status: done, write the Lessons section, run roll-up.
Quick Reference
| Step | Action | Tool |
|---|---|---|
| Decide durable | Check .dojo/delegation.yaml triggers |
view .dojo/delegation.yaml |
| Create task | New board file from template | powershell → bash scripts/board.sh new "<title>" |
| Status snapshot | Group tasks by status | powershell → bash scripts/board.sh list |
| Single-line summary | Counts per status | powershell → bash scripts/board.sh status |
| Update todo.md | Roll up board into the plan view | powershell → bash scripts/board.sh roll-up |
| Edit task fields | Update frontmatter or checkboxes | edit tasks/board/<file>.md |
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.
- 9d ago First seen · 105 lines · 17 tokens per session scan A 04d93e279965
durable-work is a skill published in the GitHub repository andreaswasita/copilot-agents-dojo (52 stars, last pushed yesterday), licensed MIT. It adds 17 tokens to every session and 1,327 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
starting-a-new-project
Use when the workspace is empty — no code yet — and the user brings a raw idea: the brand-new branch of setting-up-a-project, normally reached via that dispatcher, directly only when the situation is unmistakable. Not for features in an existing project — use brainstorming instead.
file-a-task
File work into nohuman (taskadd) and check on it (taskstatus) via the nohuman MCP bridge, instead of doing the work inline.
client-onboarding
Use when a deal closed or a user signed up and the first 30 days need an activation plan: sales→delivery handoff, one verifiable activation event, kickoff, and a 30/60/90 or day-0→14 plan with owners, dates and a measurable exit. NOT reactive ticket triage (that is customer-support), NOT renewals/churn past the…
grants
Use when pursuing non-dilutive public or foundation funding — money you neither repay nor give equity for: finding a fit call, go/no-go, registration (UEI/PIC/BDNS), or writing the rubric-scored application, logic model, budget justification, or LOI. NOT an equity round (that is fundraising), NOT a paying-customer SOW…
init
Use when starting from nothing or pointing rsc at an existing project — the front door. Gauges technical level and the accompaniment dial first (non-technical by default), discovers what the user wants to build or govern (any stack, or a non-code harness: company/ops, research, knowledge, content), writes the profile…
no-code-app
Use when building an app on a no-code/low-code platform (Bubble, Softr, Glide over Airtable or Xano) — picking the platform, modeling the data, keeping usage-metered cost predictable, and planning the exit before lock-in. NOT a hand-written coded app (that is nextjs), NOT wiring SaaS tools together with no app UI…