Paper Context Resolver
Resolve a narrow, reproduction-critical detail from primary paper sources when README and repo files leave a documented gap.
Overview
paper-context-resolver is an agent skill for the Build phase that resolves narrow reproduction-critical paper details and records conflicts when README and repository files are incomplete.
Install
npx skills add https://github.com/lllllllama/rigorpilot-skills --skill paper-context-resolverWhat is this skill?
- Optional helper-tier skill—usually orchestrated by ai-research-reproduction
- Targets narrow gaps: dataset split, preprocessing, eval protocol, checkpoint mapping, runtime assumptions
- Pulls evidence from primary paper sources and records conflicts with README
- Requires a concrete reproduction question—not title-only paper lookup
- Does not override README instructions without documenting the conflict
Adoption & trust: 32.3k installs on skills.sh; 412 GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
Reproduction is blocked because the README and repo omit one protocol detail that only the paper clarifies, and you need evidence—not a generic summary.
Who is it for?
Targeted paper lookups during active DL repo reproduction when intake already identified a single critical ambiguity.
Skip if: General paper summaries, replacing README-first workflow, title-only searches, or overriding repo docs without recording the disagreement.
When should I use this skill?
README and repository files leave a narrow reproduction-critical gap and there is a concrete question about paper-grounded protocol detail.
What do I get? / Deliverables
You get a documented answer to the specific reproduction gap with primary-source citations and explicit README or repo conflict notes for the repro_outputs trail.
- Documented paper-sourced answer
- Conflict notes versus README or repo
Recommended Skills
Journey fit
Paper evidence supplements active repo reproduction during build, not generic literature review at the idea phase. integrations because the skill bridges paper protocol details (splits, checkpoints, eval) into the running reproduction stack without replacing README-first guidance.
How it compares
Narrow reproduction evidence supplement—not a standalone literature-review or repo-scan skill.
Common Questions / FAQ
Who is paper-context-resolver for?
Solo builders running RigorPilot reproduction who need one paper-backed fact to align commands, data splits, or checkpoints with the published method.
When should I use paper-context-resolver?
During build-phase reproduction after README and repo review, when a concrete gap remains about splits, preprocessing, eval protocol, checkpoint mapping, or runtime assumptions—and not when the README already suffices.
Is paper-context-resolver safe to install?
It focuses on read-only research resolution and conflict logging; review the Security Audits panel on this page and note network access may be needed to fetch paper sources.
Workflow Chain
Requires first: ai research reproduction
SKILL.md
READMESKILL.md - Paper Context Resolver
# paper-context-resolver Use this as the Rigor Paper Context helper. The installed slug remains `paper-context-resolver` for compatibility. ## When to apply - README and repo files leave a reproduction-critical gap. - The gap concerns dataset version, split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumptions. - The main skill needs a narrow evidence supplement instead of a full paper summary. - There is already a concrete reproduction question to answer. ## When not to apply - The README already gives enough reproduction detail. - The user wants a general paper explanation rather than reproduction support. - The goal is to override README instructions without documenting the conflict. - The only available input is a paper title and there is no concrete reproduction gap yet. ## Clear boundaries - This skill is optional. - This skill is helper-tier and should usually be orchestrator-invoked. - It supplements README-first reproduction. - It does not replace the main orchestration flow. - It does not summarize the whole paper by default. ## Input expectations - target repo metadata - reproduction-critical question - existing README or repo evidence - any already known paper links ## Output expectations - narrowed source list - reproduction-relevant answer only - explicit README-paper conflict note when applicable - clear distinction between direct evidence and inference ## Notes Use `references/paper-assisted-reproduction.md`. # Paper-Assisted Reproduction ## Purpose Use paper context only to unblock README-first reproduction when the repository leaves a critical gap. Only invoke this step after README and repo inspection already produced a concrete unanswered question. ## Source order 1. paper link explicitly provided by the README 2. official project page or official paper page 3. arXiv or OpenReview primary paper record 4. Google Scholar only to help locate the primary source ## Allowed question types - dataset version or split - preprocessing or postprocessing details - evaluation protocol details - checkpoint or model variant mapping - critical runtime assumptions ## Trigger discipline Good trigger: - "README is ambiguous about the evaluation split. Check the linked paper and record the conflict if needed." Bad trigger: - "Summarize this paper." - "Find the paper for this repository title and explain it." - "Tell me everything the paper changed compared with prior work." ## Default exclusions - full paper summary - novelty explanation - broad related-work discussion - speculative "best settings" not tied to reproduction ## Conflict rule If README and paper disagree: - do not silently replace README - record the conflict - explain which source says what - preserve the README-first policy in the final report display_name: Rigor Paper Context short_description: Rigor Paper Context helper for narrow reproduction-critical paper gaps without overriding README by default. default_prompt: Use the linked paper only to resolve a narrow reproduction-critical detail such as dataset split, evaluation protocol, preprocessing, or checkpoint mapping, and record any README conflicts.