fix startup packaging and content guardrails
This commit is contained in:
Advait Paliwal
@@ -5,7 +5,7 @@ description: Autonomous experiment loop that tries ideas, measures results, keep
|
||||
|
||||
# Autoresearch
|
||||
|
||||
Run the `/autoresearch` workflow. Read the prompt template at `prompts/autoresearch.md` for the full procedure.
|
||||
Run the `/autoresearch` workflow. Read the prompt template at `../prompts/autoresearch.md` for the full procedure.
|
||||
|
||||
Tools used: `init_experiment`, `run_experiment`, `log_experiment` (from pi-autoresearch)
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Contribute changes to the Feynman repository itself. Use when the t
|
||||
|
||||
# Contributing
|
||||
|
||||
Read `CONTRIBUTING.md` first, then `AGENTS.md` for repo-level agent conventions.
|
||||
Read `../CONTRIBUTING.md` first, then `../AGENTS.md` for repo-level agent conventions.
|
||||
|
||||
Use this skill when working on Feynman itself, especially for:
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Run a thorough, source-heavy investigation on any topic. Use when t
|
||||
|
||||
# Deep Research
|
||||
|
||||
Run the `/deepresearch` workflow. Read the prompt template at `prompts/deepresearch.md` for the full procedure.
|
||||
Run the `/deepresearch` workflow. Read the prompt template at `../prompts/deepresearch.md` for the full procedure.
|
||||
|
||||
Agents used: `researcher`, `verifier`, `reviewer`
|
||||
|
||||
|
||||
@@ -5,6 +5,6 @@ description: Inspect active background research work including running processes
|
||||
|
||||
# Jobs
|
||||
|
||||
Run the `/jobs` workflow. Read the prompt template at `prompts/jobs.md` for the full procedure.
|
||||
Run the `/jobs` workflow. Read the prompt template at `../prompts/jobs.md` for the full procedure.
|
||||
|
||||
Shows active `pi-processes`, scheduled `pi-schedule-prompt` entries, and running subagent tasks.
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Run a literature review using paper search and primary-source synth
|
||||
|
||||
# Literature Review
|
||||
|
||||
Run the `/lit` workflow. Read the prompt template at `prompts/lit.md` for the full procedure.
|
||||
Run the `/lit` workflow. Read the prompt template at `../prompts/lit.md` for the full procedure.
|
||||
|
||||
Agents used: `researcher`, `verifier`, `reviewer`
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Compare a paper's claims against its public codebase. Use when the
|
||||
|
||||
# Paper-Code Audit
|
||||
|
||||
Run the `/audit` workflow. Read the prompt template at `prompts/audit.md` for the full procedure.
|
||||
Run the `/audit` workflow. Read the prompt template at `../prompts/audit.md` for the full procedure.
|
||||
|
||||
Agents used: `researcher`, `verifier`
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Turn research findings into a polished paper-style draft with secti
|
||||
|
||||
# Paper Writing
|
||||
|
||||
Run the `/draft` workflow. Read the prompt template at `prompts/draft.md` for the full procedure.
|
||||
Run the `/draft` workflow. Read the prompt template at `../prompts/draft.md` for the full procedure.
|
||||
|
||||
Agents used: `writer`, `verifier`
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Simulate a tough but constructive peer review of an AI research art
|
||||
|
||||
# Peer Review
|
||||
|
||||
Run the `/review` workflow. Read the prompt template at `prompts/review.md` for the full procedure.
|
||||
Run the `/review` workflow. Read the prompt template at `../prompts/review.md` for the full procedure.
|
||||
|
||||
Agents used: `researcher`, `reviewer`
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Plan or execute a replication of a paper, claim, or benchmark. Use
|
||||
|
||||
# Replication
|
||||
|
||||
Run the `/replicate` workflow. Read the prompt template at `prompts/replicate.md` for the full procedure.
|
||||
Run the `/replicate` workflow. Read the prompt template at `../prompts/replicate.md` for the full procedure.
|
||||
|
||||
Agents used: `researcher`
|
||||
|
||||
|
||||
@@ -5,6 +5,6 @@ description: Write a durable session log capturing completed work, findings, ope
|
||||
|
||||
# Session Log
|
||||
|
||||
Run the `/log` workflow. Read the prompt template at `prompts/log.md` for the full procedure.
|
||||
Run the `/log` workflow. Read the prompt template at `../prompts/log.md` for the full procedure.
|
||||
|
||||
Output: session log in `notes/session-logs/`.
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Compare multiple sources on a topic and produce a grounded comparis
|
||||
|
||||
# Source Comparison
|
||||
|
||||
Run the `/compare` workflow. Read the prompt template at `prompts/compare.md` for the full procedure.
|
||||
Run the `/compare` workflow. Read the prompt template at `../prompts/compare.md` for the full procedure.
|
||||
|
||||
Agents used: `researcher`, `verifier`
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ description: Set up a recurring research watch on a topic, company, paper area,
|
||||
|
||||
# Watch
|
||||
|
||||
Run the `/watch` workflow. Read the prompt template at `prompts/watch.md` for the full procedure.
|
||||
Run the `/watch` workflow. Read the prompt template at `../prompts/watch.md` for the full procedure.
|
||||
|
||||
Agents used: `researcher`
|
||||
|
||||
|
||||
Reference in New Issue
Block a user