engram-frame
Turn a wish into a brief the factory can run on. YOU are the orchestrator
holding this skill — no child agent has the owner channel. You interview the
owner (a Product Engineer, technical or functional), capture intent, never
design. The brief is the human-to-factory contract: everything the analyzer
would otherwise guess.
The failure this skill exists to prevent: a factory job where consequential
choices are decided downstream with no recorded human decision, discovered by
the human only at the approval gate.
Contract
- Capture intent, not design — no solutioning in the brief.
- The brief is functional: what the system must do, for whom, and why it
matters — language a functional reader can verify. Implementation belongs
to analysis and plan, even for technical owners.
- The owner speaks in quotes; decisions are
[HUMAN-DECIDED] only with an
explicit selection or verbatim statement. Silence is never consent.
- Every consequential choice lands in the decision ledger: decided by the
human, or explicitly delegated with a boundary and a revisit trigger.
- The ask is the owner's words, verbatim, any length.
- Stop interviewing when a different answer would no longer change what
analysis does.
Question format (use everywhere)
Every question is easy to answer:
<short stem> — A) <option> B) <option> C) <option> D) freeform
- Options are grounded in repo context and prior answers, never generic
filler. Options describe behavior and outcomes, not mechanisms.
- Translate-and-confirm technical answers: when the owner answers with a
mechanism ("add a queue"), restate the functional intent and confirm: "the
behavior you want is X — is the mechanism fixed, or a suggestion?" Fixed →
owner-fixed constraint; suggestion → non-binding suggestions.
- One-word answers are valid: "B", "B but capped at 500", "none — …".
- You may recommend: "I'd pick B because — pick or correct." Disclose
the assumption behind any consequential option ("option A assumes —
correct me if not"), and record the owner's REJECTION verbatim — a refusal
is intent, not noise.
- Never exceed four substantive questions per round ("anything else?" is a
transition, not a question). Caps: two rounds per area, eight rounds total;
past a cap the request is not ready — say so and stop.
Workflow
1. Classify the ambiguity
Fast path when the request names a problem, a checkable outcome, a boundary,
and its decision owner. Full path when any is missing — signals: "better /
faster / modernize / fix", no measurable outcome, unknown boundary, unidentified decisions.
2. Fast path — one confirm-and-ratify turn
Restate problem and outcome in one sentence each; ask the owner to confirm
(zero clarifiers is valid). If a named ambiguity would change analysis, ask
it in the question format — at most two. The owner's confirmation of the
restatement doubles as the whole-brief ratification — one turn, no extra
ceremony. Fill the ledger, write the brief, done.
3. Full path — interactive interview
One area per round, in this order; ≤ 4 option-based questions per round;
each area closes with "anything else?" and moves on. Hesitation on any
choice → park it [OPEN] immediately; never push a third time.
- Problem — offer candidate framings from context:
"Which hurts most? A) … B) … C) several — rank them."
- Outcome — offer candidate measures:
"How do we know it worked? A) B) C) owner-judged pass/fail
(tell me the signal) D) freeform."
- Scope by MoSCoW — propose the buckets, owner edits:
Must: <items> Should: <items>
Could: <items> Won't: <items>
"Keep as-is, or move items / strike lines." Must → outcome core;
Should → in scope; Could → follow-ups; Won't → non-goals. One round, no
item-by-item interrogation.
- Boundaries by strike-list — never blank-page. Propose what you believe
is OUT (from repo context): "Strike any that should stay IN; add what's
missing." Same for constraints: propose the known standards, owner edits.
- Decisions as tradeoff pairs — each consequential choice as a
functional either/or, the tradeoff stated in behavior terms:
"A) B)
C) delegate to analysis within D) leave OPEN."
The owner's pick — including C or D — is the ledger row. C and D require
an explicit selection: never record DELEGATED because the owner hesitated
or because you judged the choice mechanical. Without a pick: OPEN, which
blocks analysis.
4. Write the brief
Fill the factory brief template
(.agents/skills/engram-factory/assets/templates/brief.md). The ask is the
owner's words verbatim, any length. Ledger rows record the selection
("owner picked B: ") or the verbatim freeform, each traceable
to a retrievable owner message (channel + reference). Capture the owner's
ratification of the whole brief verbatim (actor, channel, message reference,
date) — the approval gate displays it back to the human as the forgery
control.
5. Hand off to the factory
Fail-closed provenance gate: every HUMAN-DECIDED row and the ratification
resolve to retained owner messages (quote, retrievable reference, option
context); the owner ratifies the finished brief whole — a fast-path
confirmation covers the complete brief. Missing evidence or material OPEN
rows = not ready; stop.
The brief's ask becomes the feature file. Give the orchestrator both paths:
factory.py create --feature-file <ask> --brief-file <brief>. The controller
registers the brief as an immutable, REQUIRED analysis input; analysis honors
its ledger. The interview is over; the factory asks questions now, not you.
When it fails
- Owner unavailable → tag open decisions
[BLOCKED: owner needed] and stop.
- Two owners disagree → record both verbatim under one open decision; never average.
- No checkable outcome and no owner signal → outcome stays OPEN; the brief is
NOT factory-ready. Request changes mid-interview → restart that area only.
- More rounds than the caps allow → not ready; stop.
Anti-patterns
- Blank-page questions ("what do you want?") — always offer grounded options.
- Solutioning in the brief — including an owner's own mechanism recorded as
intent instead of translated to behavior or parked as a suggestion.
- Owner-fixed constraints without an external source — internal preferences
are suggestions or delegations, never constraints.
- Recording DELEGATED on hesitation, or downgrading an OPEN row without a
verbatim owner-approved deferral in the row.
- A technical owner turning the brief into a design doc: if a statement names
how instead of what, it does not belong here.
- Interrogation drift: more than four questions a round, or a third push on a
parked choice.
- Shortening or paraphrasing the ask — quoted owner text is never compressed.
- Inventing outcome measures, selections, or ratification quotes.
- Marking
[DELEGATED] without a boundary and revisit trigger.
- Converting an
[OPEN] row into a decision to make the brief look complete.
Final check
- Did every question offer options or freeform — no blank-page prompts?
- Is every statement functional — behavior, outcome, boundary — with
mechanisms only as owner-fixed constraints or parked suggestions?
- Does every ledger row carry real authority — a recorded pick/quote, or an
explicit delegation with a boundary?
- Could the analyzer start tomorrow morning without guessing at intent?
- Is the ask still the owner's words, not yours?
1---2name: engram-frame3description: ORCHESTRATOR-run skill — the main session conducts this interactive guided interview with the owner BEFORE starting an engram-factory job; a child agent cannot hold the owner channel. Option-based questions (pick a letter or freeform), MoSCoW prioritization, strike-lists over blank-page questions. Captures problem, outcome measures, non-goals, and a decision ledger with explicit authority. Trigger on any wish that is not yet a spec ("I want X", "make X better", "fix X"), or before /engram-factory when intent, outcome, or decision authority is unclear.4metadata:5 source: https://skillmd.com/skills/majiayu000/engram-frame6---78# engram-frame910Turn a wish into a brief the factory can run on. YOU are the orchestrator11holding this skill — no child agent has the owner channel. You interview the12owner (a Product Engineer, technical or functional), capture intent, never13design. The brief is the human-to-factory contract: everything the analyzer14would otherwise guess.1516The failure this skill exists to prevent: a factory job where consequential17choices are decided downstream with no recorded human decision, discovered by18the human only at the approval gate.1920## Contract21221. Capture intent, not design — no solutioning in the brief.232. **The brief is functional**: what the system must do, for whom, and why it24 matters — language a functional reader can verify. Implementation belongs25 to analysis and plan, even for technical owners.263. The owner speaks in quotes; decisions are `[HUMAN-DECIDED]` only with an27 explicit selection or verbatim statement. Silence is never consent.284. Every consequential choice lands in the decision ledger: decided by the29 human, or explicitly delegated with a boundary and a revisit trigger.305. The ask is the owner's words, verbatim, any length.316. Stop interviewing when a different answer would no longer change what32 analysis does.3334## Question format (use everywhere)3536Every question is easy to answer:3738```39<short stem> — A) <option> B) <option> C) <option> D) freeform40```4142- Options are grounded in repo context and prior answers, never generic43 filler. **Options describe behavior and outcomes, not mechanisms.**44- **Translate-and-confirm technical answers**: when the owner answers with a45 mechanism ("add a queue"), restate the functional intent and confirm: "the46 behavior you want is X — is the mechanism fixed, or a suggestion?" Fixed →47 owner-fixed constraint; suggestion → non-binding suggestions.48- One-word answers are valid: "B", "B but capped at 500", "none — …".49- You may recommend: "I'd pick B because <reason> — pick or correct." Disclose50 the assumption behind any consequential option ("option A assumes <X> —51 correct me if not"), and record the owner's REJECTION verbatim — a refusal52 is intent, not noise.53- Never exceed four substantive questions per round ("anything else?" is a54 transition, not a question). Caps: two rounds per area, eight rounds total;55 past a cap the request is not ready — say so and stop.5657## Workflow5859### 1. Classify the ambiguity6061Fast path when the request names a problem, a checkable outcome, a boundary,62and its decision owner. Full path when any is missing — signals: "better /63faster / modernize / fix", no measurable outcome, unknown boundary, unidentified decisions.6465### 2. Fast path — one confirm-and-ratify turn6667Restate problem and outcome in one sentence each; ask the owner to confirm68(zero clarifiers is valid). If a named ambiguity would change analysis, ask69it in the question format — at most two. The owner's confirmation of the70restatement doubles as the whole-brief ratification — one turn, no extra71ceremony. Fill the ledger, write the brief, done.7273### 3. Full path — interactive interview7475One area per round, in this order; ≤ 4 option-based questions per round;76each area closes with "anything else?" and moves on. Hesitation on any77choice → park it `[OPEN]` immediately; never push a third time.78791. **Problem** — offer candidate framings from context:80 "Which hurts most? A) … B) … C) several — rank them."812. **Outcome** — offer candidate measures:82 "How do we know it worked? A) <metric> B) <signal> C) owner-judged pass/fail83 (tell me the signal) D) freeform."843. **Scope by MoSCoW** — propose the buckets, owner edits:85 ```86 Must: <items> Should: <items>87 Could: <items> Won't: <items>88 ```89 "Keep as-is, or move items / strike lines." Must → outcome core;90 Should → in scope; Could → follow-ups; Won't → non-goals. One round, no91 item-by-item interrogation.924. **Boundaries by strike-list** — never blank-page. Propose what you believe93 is OUT (from repo context): "Strike any that should stay IN; add what's94 missing." Same for constraints: propose the known standards, owner edits.955. **Decisions as tradeoff pairs** — each consequential choice as a96 functional either/or, the tradeoff stated in behavior terms:97 "A) <reversible behavior> B) <faster but riskier behavior>98 C) delegate to analysis within <boundary> D) leave OPEN."99 The owner's pick — including C or D — is the ledger row. **C and D require100 an explicit selection**: never record DELEGATED because the owner hesitated101 or because you judged the choice mechanical. Without a pick: OPEN, which102 blocks analysis.103104### 4. Write the brief105106Fill the factory brief template107(`.agents/skills/engram-factory/assets/templates/brief.md`). The ask is the108owner's words verbatim, any length. Ledger rows record the selection109("owner picked B: <option text>") or the verbatim freeform, each traceable110to a retrievable owner message (channel + reference). Capture the owner's111ratification of the whole brief verbatim (actor, channel, message reference,112date) — the approval gate displays it back to the human as the forgery113control.114115### 5. Hand off to the factory116117Fail-closed provenance gate: every HUMAN-DECIDED row and the ratification118resolve to retained owner messages (quote, retrievable reference, option119context); the owner ratifies the finished brief whole — a fast-path120confirmation covers the complete brief. Missing evidence or material OPEN121rows = not ready; stop.122123The brief's ask becomes the feature file. Give the orchestrator both paths:124`factory.py create --feature-file <ask> --brief-file <brief>`. The controller125registers the brief as an immutable, REQUIRED analysis input; analysis honors126its ledger. The interview is over; the factory asks questions now, not you.127128## When it fails129130- Owner unavailable → tag open decisions `[BLOCKED: owner needed]` and stop.131- Two owners disagree → record both verbatim under one open decision; never average.132- No checkable outcome and no owner signal → outcome stays OPEN; the brief is133 NOT factory-ready. Request changes mid-interview → restart that area only.134- More rounds than the caps allow → not ready; stop.135136## Anti-patterns137138- Blank-page questions ("what do you want?") — always offer grounded options.139- Solutioning in the brief — including an owner's own mechanism recorded as140 intent instead of translated to behavior or parked as a suggestion.141- Owner-fixed constraints without an external source — internal preferences142 are suggestions or delegations, never constraints.143- Recording DELEGATED on hesitation, or downgrading an OPEN row without a144 verbatim owner-approved deferral in the row.145- A technical owner turning the brief into a design doc: if a statement names146 how instead of what, it does not belong here.147- Interrogation drift: more than four questions a round, or a third push on a148 parked choice.149- Shortening or paraphrasing the ask — quoted owner text is never compressed.150- Inventing outcome measures, selections, or ratification quotes.151- Marking `[DELEGATED]` without a boundary and revisit trigger.152- Converting an `[OPEN]` row into a decision to make the brief look complete.153154## Final check155156- Did every question offer options or freeform — no blank-page prompts?157- Is every statement functional — behavior, outcome, boundary — with158 mechanisms only as owner-fixed constraints or parked suggestions?159- Does every ledger row carry real authority — a recorded pick/quote, or an160 explicit delegation with a boundary?161- Could the analyzer start tomorrow morning without guessing at intent?162- Is the ask still the owner's words, not yours?