Menu
Skills dans ce dépôt
SkillsMP a collecté 72 skills depuis WeaveMindAI/weft. Ouvrez un skill pour examiner sa source et ses détails.
WeaveMindAI/weftAffichage de 32 skills collectés sur 72.
COMMAND, not reference: Grow the program one stage at a time against a real input, seeded runs and frozen examples included. Run this when the user asks for this step by name, optionally naming the stage to grow or an example name to rerun. The sixteen…
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
COMMAND, not reference: Grow the program one stage at a time against a real input, seeded runs and frozen examples included. Run this when the user asks for this step by name, optionally naming the stage to grow or an example to check. The sixteen `weft-`…
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Writes and overhauls the LLM prompts inside a weft program (system prompts in prompts/, user prompt templates, JSON extraction instructions). Runs on the WeaveMind prompt-building playbook. Dispatched by Tangle when a stage talks to a model and the prompt…
Langue du texte source : anglais
Attacks a weft program before it faces the world, hunting every credible hole: an outsider lying through an input, a hallucination the graph trusts, a rogue step with agency, an unguarded path to something that matters. Dispatched by Tangle when the program…
Langue du texte source : anglais
Deep post-mortem digging into a weft execution. Dispatched with a color or a symptom when the cause is unclear or the journals are long; reconstructs exactly what happened from events, logs, source, node code, and stored files, compares good runs against bad…
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Writes and overhauls the LLM prompts inside a weft program (system prompts in prompts/, user prompt templates, JSON extraction instructions). Runs on the WeaveMind prompt-building playbook. Dispatched by Tangle when a stage talks to a model and the prompt…
Langue du texte source : anglais
Attacks a weft program before it faces the world, hunting every credible hole: an outsider lying through an input, a hallucination the graph trusts, a rogue step with agency, an unguarded path to something that matters. Dispatched by Tangle when the program…
Langue du texte source : anglais
Deep post-mortem digging into a weft execution. Dispatched with a color or a symptom when the cause is unclear or the journals are long; reconstructs exactly what happened from events, logs, source, node code, and stored files, compares good runs against bad…
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Read when the user asks to update or upgrade weft, or when something broke right after one: locating the checkout, the git pull plus setup.sh walk, what setup.sh rebuilds and what it preserves, and fixing a failed update.
Langue du texte source : anglais
Load full codebase context before we design a feature
Langue du texte source : anglais
Write the agreed design to a plan file in ~/.codex/plans/
Langue du texte source : anglais
Fully implement the agreed plan in one long autonomous session
Langue du texte source : anglais
One wide-net review round over the whole uncommitted diff, then clean reshape fixes
Langue du texte source : anglais
Paranoid red-team pass over the fixes made during review, re-verifying each one
Langue du texte source : anglais
Babble mode: stream of consciousness, no structure, half-thoughts and dead ends, thinking out loud. Load when switching to [babble mode] or when early exploration should run unstructured before convergence.
Langue du texte source : anglais
Collaborative mode, the default cognitive pattern: build on ideas, riff, propose, keep forward momentum. Load when switching to [collaborative mode] or when the conversation is open-ended design work.
Langue du texte source : anglais
Convergence mode: synthesize scattered thoughts into coherent structure, extract core insights, output actionable next steps. Load when switching to [convergence mode] or when a rambling discussion needs pulling together.
Langue du texte source : anglais
Debug mode: observe, hypothesize, test, evaluate, fix-and-stop; no certainty without evidence, no symptom patches. Load when switching to [debug mode] or when diagnosing a bug or unexpected behavior.
Langue du texte source : anglais
Red team mode: systematically challenge every assumption, hunt blind spots, output counter-perspectives after each claim. Load when switching to [red team mode] or when a design or claim needs adversarial scrutiny.
Langue du texte source : anglais
Research mode: search before implementing, search for concepts not libraries, overdo it, never trust stale knowledge. Load when switching to [research mode] or when blocked on a design decision, library choice, or anything the internet answers.
Langue du texte source : anglais
Writing mode: distill prose, fight length and jargon, loop drafts through the AI-tell catalog until a full pass finds nothing. Load when switching to [writing mode] or before writing any substantial prose (docs, READMEs, design notes, journal entries, long…
Langue du texte source : anglais
Writing mode: distill prose, fight length and jargon, loop drafts through the AI-tell catalog until a full pass finds nothing. Load when switching to [writing mode] or before writing any substantial prose (docs, READMEs, design notes, journal entries, long…
Langue du texte source : anglais
Collaborative mode, the default cognitive pattern: build on ideas, riff, propose, keep forward momentum. Load when switching to [collaborative mode] or when the conversation is open-ended design work.
Langue du texte source : anglais
Writing mode: distill prose, fight length and jargon, loop drafts through the AI-tell catalog until a full pass finds nothing. Load when switching to [writing mode] or before writing any substantial prose (docs, READMEs, design notes, journal entries, long…
Langue du texte source : anglais