CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

paper2code

Converts an arxiv paper into a minimal, citation-anchored Python implementation. Trigger when user runs /paper2code with an arxiv URL or paper ID, says "implement this paper", or pastes an arxiv link asking for implementation. Flags all ambiguities honestly. Never invents implementation details not stated in the paper.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/paper2code/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured orchestration file with a clear staged workflow, concrete commands, and good token discipline, undermined by an incomplete bundle: the pipeline, guardrails, knowledge, and scaffold files it dispatches to do not exist, so most stages reduce to dangling references. Script invocation paths also assume an install layout (skills/paper2code/scripts/) that differs from the bundle layout (scripts/).

Suggestions

Ship the referenced bundle files (pipeline/01-05, guardrails/, knowledge/, scaffolds/) or inline their critical content — as bundled, every stage past acquisition dispatches to a file that does not exist.

Make script paths match the bundle layout (e.g., `python scripts/fetch_paper.py ...` relative to the skill root) so the commands run as written instead of assuming a skills/paper2code/ install location.

Add explicit validation checkpoints to Stages 2-5 (e.g., confirm contribution.md and ambiguity_audit.md were written before proceeding, and run the notebook sanity checks before cleanup) to match the checkpoint quality already present in Stage 1.

DimensionReasoningScore

Conciseness

The body is a lean orchestrator: no concept explanations, concrete commands, terse mode definitions — "Core contribution only. Training loop only if contribution involves training." Not 5 because of mild redundancy: guardrail files are referenced twice (in Stages 3-4 and again in a dedicated section) and the official-code paragraph in Stage 1 could be trimmed.

4 / 5

Actionability

Stages 1's commands are copy-paste ready (pip install, fetch_paper.py, extract_structure.py) and artifact names are specific (contribution.md, ambiguity_audit.md, walkthrough.ipynb), but Stages 2, 3, and 5 provide only one-line summaries and dispatch entirely to pipeline/*.md files that are absent from the bundle, leaving key execution details missing. Not 4 because the guidance is not executable end-to-end as shipped; not 2 because the acquisition stage and artifact contracts are concrete.

3 / 5

Workflow Clarity

Clear ordered sequence ("Do NOT skip stages. Do NOT combine stages. Execute them in order.") with one explicit validation checkpoint and feedback loop ("Verify the outputs exist before proceeding. If extraction failed, follow the fallback protocol") plus link verification and post-notebook sanity checks. Not 5 because Stages 2-5 lack explicit checkpoints (e.g., no verify-that-artifact-was-written step before proceeding).

4 / 5

Progressive Disclosure

The design is textbook progressive disclosure — a short orchestration body with well-signaled, one-level-deep references to pipeline/, guardrails/, knowledge/, and scaffolds/ — but scored against the actual bundle, only scripts/ exists: all five pipeline files, three guardrail files, four knowledge files, and the scaffolds are missing, so navigation breaks at every stage. Not 4 because the broken references are a major, not minor, organization gap given the whole skill depends on them; not 2 because the structure that is present (stages, scripts, section headers) is clear and not buried.

3 / 5

Total

14

/

20

Passed

Description

82%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: third-person voice, explicit what-and-when structure with concrete trigger phrases, and a well-differentiated niche. The only notable gap is capability coverage — several concrete actions the skill performs (parsing, notebooks, modes, frameworks) are omitted, keeping specificity at mid-range.

DimensionReasoningScore

Specificity

"Converts an arxiv paper into a minimal, citation-anchored Python implementation" plus "Flags all ambiguities honestly" names the domain and 1-2 concrete actions, but coverage has gaps — parsing, walkthrough notebooks, modes (minimal/full/educational), and framework options are never mentioned. Not 4 because anchor 4 expects several specific listed actions; not 2 because the actions given are concrete rather than generic.

3 / 5

Completeness

Clearly answers both: what ("Converts an arxiv paper into a minimal, citation-anchored Python implementation") and when ("Trigger when user runs /paper2code with an arxiv URL or paper ID, says 'implement this paper', or pastes an arxiv link asking for implementation") with concrete trigger phrases. Matches the anchor-5 example structure exactly; nothing above it exists.

5 / 5

Trigger Term Quality

Includes natural phrases users would actually say — "implement this paper", "arxiv URL", "paper ID", "pastes an arxiv link" — plus the /paper2code command. Not 5 because common synonyms like "reproduce this paper", "code this paper up", or arxiv.org/.abs forms are missing.

4 / 5

Distinctiveness Conflict Risk

Clear niche (arxiv paper → Python implementation) with distinct triggers including a dedicated slash command and arxiv-specific anchoring. Minimal conflict risk with generic coding or document skills; not 4 because there is no meaningful overlap with closely related skills.

5 / 5

Total

17

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

PrathamLearnsToCode/paper2code

Path

skills/paper2code/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.