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/DailybotHQ/deepworkplan-skillnpx agentmods add skills/dailybothq/deepworkplan-skill/upgradeWrote 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/dailybothq/deepworkplan-skill/upgrade)<a href="https://agentmods.dev/skills/dailybothq/deepworkplan-skill/upgrade"><img src="https://agentmods.dev/badge/skills/dailybothq/deepworkplan-skill/upgrade/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/dailybothq/deepworkplan-skill/upgrade"><img src="https://agentmods.dev/badge/skills/dailybothq/deepworkplan-skill/upgrade.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00143 | $0.02895 |
| Opus 5.5 | $0.00057 | $0.01158 |
| Sonnet 5.5 | $0.00029 | $0.00579 |
| Haiku 4.5 | $0.00014 | $0.00290 |
Grade A, and why
deepworkplan-upgrade 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 — 199 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DeepWorkPlan — Upgrade (Sub-Skill)
Keep a repository's installed DeepWorkPlan skill and its declared DWP standard
current. Three phases, strictly ordered: check (read-only, safe to run
anytime), consent (nothing downloads without an explicit yes), upgrade
(same install command a fresh consumer runs, then onboarding re-executed as if
https://deepworkplan.com/init.md were executed fresh). An upgrade never touches
.dwp/ plans — plans and their recorded evidence are history, not state to
migrate (../spec/DWP_SPECIFICATION.md §6.5); its one .dwp/ write is the
addon registry back-fill (Phase 3 step 4).
Trust boundary (write scope)
allowed-tools includes write-capable Edit, Write and Bash. Phase 1
(check) and Phase 2 (consent) are read-only. Only after the developer's
explicit acceptance does Phase 3 write: the vendored skill tree at the exact
accepted stable tag (pre-release tags only on an explicit request for that
channel), through the documented installer, then the onboarding
reconciliation. It MUST NOT download anything before acceptance, install
a moving ref (@main, @latest) or an unpinned source, overwrite local
adaptations without showing their diff first, touch any plan under .dwp/,
or read, print or write a secret.
When to use
- "upgrade DWP", "update the deepworkplan skill", "is there a newer version?", "re-run the init", "/dwp-upgrade"
- Before a long plan, when the developer wants the harness current
- Not for first-time onboarding (route to
../onboard/SKILL.md) - Not for changing an in-flight plan's scope or shape (
../refine/SKILL.md) - Not for the targeted harness upgrade of one outdated piece — that is
../onboard/SKILL.mdPhase 0; this sub-skill replaces the whole skill first
Context detection
Run the shared context helper to resolve the repo, branch, agent and .dwp/
location:
bash ../shared/context.sh
Then locate the installed pack: the deepworkplan/ directory this sub-skill
was read from (in a repo, typically .agents/skills/deepworkplan/; a global
install lives under ~/.<agent>/skills/deepworkplan/). Everything below reads
versions relative to that directory.
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 Changed 3bc94d74f876
- 4d ago Changed c06cd7f8f8cb
- 5d ago Changed 1da412668870
- 6d ago Changed d2fde02092f1
- 7d ago Changed · +47 lines 3868af530fa2
- 9d ago Changed 0ba103292e4a
- 11d ago Changed 1a0cb7a4d67d
- 15d ago Changed 718c73c24ef9
- 18d ago Changed · -1 lines 7cae8fe0b547
- 29d ago First seen · 153 lines · 143 tokens per session scan A be7d38bc8e58
deepworkplan-upgrade is a skill published in the GitHub repository DailybotHQ/deepworkplan-skill (23 stars, last pushed 3d ago), licensed MIT. It adds 143 tokens to every session and 2,895 once invoked, about $0.0006 per session on Opus 5.5. 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 2026-09-13.
Other skills, from other repositories
gsd-plan-review-convergence
Cross-AI plan convergence - replan until review concerns are resolved.
gsd-ns-manage
config workspace | workstreams thread update ship inbox.
gsd-validate-phase
Retroactively audit and fill Nyquist validation gaps for a completed phase.
dependency-upgrade
Plan, batch, and verify dependency upgrades safely. Triages outdated packages into risk tiers, upgrades in order (dev/minor/patch first, runtime majors last), verifies each batch, and produces an auditable commit sequence. Use when asked to "upgrade deps", "bump packages", "update nodemodules", "fix vulnerabilities"…
forensics
Post-mortem a failed GSD auto-mode run. Traces symptom to root cause via .gsd/ activity, journal, metrics, and lock artifacts, producing a filing-ready bug report with file:line refs and a fix suggestion. Use when asked to "forensics", "post-mortem", "why did auto-mode fail", "trace the stuck loop", "debug the crash"…
review
Review code changes for security, performance, bugs, and quality. Reviews staged changes, unstaged changes, specific commits, or PR-ready diffs.