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 skills add brunob54/superpowers-orchestrator --skill researching-prior-artgit clone --depth 1 https://github.com/brunob54/superpowers-orchestratorWrote 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/brunob54/superpowers-orchestrator/researching-prior-art)<a href="https://agentmods.dev/skills/brunob54/superpowers-orchestrator/researching-prior-art"><img src="https://agentmods.dev/badge/skills/brunob54/superpowers-orchestrator/researching-prior-art/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/brunob54/superpowers-orchestrator/researching-prior-art"><img src="https://agentmods.dev/badge/skills/brunob54/superpowers-orchestrator/researching-prior-art.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 | $0.00133 | $0.09054 |
| Opus 5 | $0.00067 | $0.04527 |
| Sonnet 5 | $0.00027 | $0.01811 |
| Haiku 4.5 | $0.00013 | $0.00905 |
Grade A, and why
researching-prior-art 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 yesterday.
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 — 623 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Researching Prior Art
Gather verified external evidence for one technology decision, in
parallel, without touching the working tree outside docs/research/ and
.superpowers/research/. Evidence only — this skill never chooses an
approach.
When invoked by brainstorming, never ask the user anything — all user interaction already happened at brainstorming's research gate. The one exception: invoked directly by the user without an explicit N, present the gate message below yourself to obtain N (the user is the invoker; no gate has happened yet). Never pick N silently.
On every invocation path — whether the gate message below is shown or
not — a non-zero N ends, on success, with findings cached under
docs/research/ and committed to the currently checked-out branch in
this repository, unless HEAD is detached, in which case the commit is
skipped and the cache stays uncommitted on disk (step 6 item 5, sub-step
5e). State this to the invoker before that commit can run.
On that direct path, present this gate message verbatim — the same
block brainstorming presents, copied character-exactly (<candidates>,
<S>, and <branch> filled in — the backticks around them are
placeholder markup, dropped with the angle brackets when the values
are filled; the bracketed sentence appears only when the choice is
difficult to reverse):
Research gate: this decision triggers prior-art research. Candidates:
<candidates>. [This choice is difficult to reverse — consider a higher N.] How many research subagents should I dispatch? Suggested N=<S>(number of candidates + 3, at most 10). Reply with a number, or 0 to skip — a skip is recorded in the spec. On a non-zero reply, findings are cached underdocs/research/and committed to this repository, on the branch checked out right now (<branch>).
A direct invocation usually has no spec. On a 0 reply, state the skip
in the conversation and stop; record it in a spec's "Prior art and
alternatives" section only when a spec exists. <branch> = the name
of the branch checked out at the moment the gate fires, resolved like
this: first run git rev-parse --git-dir at the repo root. A
non-zero exit means this is not a git repository — fill <branch>
with the text not a git repository — the cache will not be committed and stop there; do not run git symbolic-ref -q HEAD in
that case, its exit status does not distinguish a non-git project
from a detached HEAD. When git rev-parse --git-dir succeeds, next
run git symbolic-ref --short -q HEAD. A non-zero exit here means
HEAD is detached — fill <branch> with the text
detached HEAD — the cache will not be committed. A zero exit means
HEAD is attached to a branch, and the command's output is the branch
name: <branch> = that output. This works on a branch with no
commits yet (a repository right after git init). Never use
git rev-parse --abbrev-ref HEAD for this: on a detached HEAD it
prints the literal string HEAD, and on a branch with no commits it
fails and still prints HEAD, so in both cases its output looks
like a branch name and is not one. A negative or
non-numeric reply → ask once more; a second unusable reply → treat it
as a skip, exactly as a 0 reply: state the skip and stop, and never
fall back to the suggested <S>. A non-zero N dispatches researchers
and ends with a commit to the user's checked-out branch, and a user
who twice answered something other than a number has not agreed to
either. This is the same rule brainstorming applies at its gate.
What ships with it
2 files 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.
- yesterday Changed · +6 lines e2c5152b423b
- 11d ago First seen · 617 lines · 133 tokens per session scan A c06f2b4e92fa
researching-prior-art is a skill published in the GitHub repository brunob54/superpowers-orchestrator (3 stars, last pushed yesterday), licensed MIT. It adds 133 tokens to every session and 9,054 once invoked, about $0.0007 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
github-pr-review
Handles PR review comments and feedback resolution. Use when user wants to resolve PR comments, handle review feedback, fix review comments, address PR review, check review status, respond to reviewer, verify PR readiness, review PR comments, analyze review feedback, evaluate PR comments, assess review suggestions, or…
privacy-guard
Prevents private infrastructure details (node hostnames, internal project names, local usernames and personal emails, absolute home paths, private and VPN IP ranges) from leaking into public repositories through commits, PRs, docs or release artifacts. Use when working in a public or soon-to-be-public repo, before…
github-pr-merge
Merges GitHub Pull Requests after validating pre-merge checklist. Use when user wants to merge PR, close PR, finalize PR, complete merge, approve and merge, or execute merge. Runs pre-merge validation (tests, lint, CI, comments), confirms with user, merges with proper format, handles post-merge cleanup.
github-pr-creation
Creates GitHub Pull Requests with automated validation and task tracking. Use when user wants to create PR, open pull request, submit for review, or check if ready for PR. Analyzes commits, validates task completion, generates Conventional Commits title and description, suggests labels. NOTE - for merging existing…
git-commit
Creates git commits following Conventional Commits format with type/scope/subject. Use when user wants to commit changes, create commit, save work, or stage and commit. Enforces project-specific conventions from CLAUDE.md.
qa
Code review + QA vs Acceptance Criteria, including A+ status suggestions and review comment draft.