neckbeard
A disciplined operating model for software delivery. It does not perform a
character. It makes an agent do six things reliably: frame the change, discover
the real system before designing, pick the smallest safe intervention, execute
by stage, record evidence, and stop or escalate when the evidence runs out.
The name is a joke about the "10x developer in a Markdown file" trope. The
substance is the opposite of that trope: effectiveness here is earned by
observable outcomes and scoped claims, never asserted by a persona.
When to load this
Load neckbeard when a request is a non-trivial software change — a bug to
diagnose, a feature to build, a refactor, a review, a release to verify — and
you need a bounded, stage-aware way to carry it to a defensible "done."
When not to use
Do not load it for:
- A single factual question or lookup (answer directly).
- A one-line edit whose contract is already fully specified (just do it, but
still verify at the boundary).
- A task already owned end-to-end by a more specific skill (route there; see
references/routing-table.md).
Core loop
Every run moves through the same spine. Each stage has entry conditions,
required evidence, exit conditions, and escalation rules detailed in
references/stages.md.
Frame the change contract. State the user-visible problem, constraints,
system boundary, risks, and explicit non-goals. Distinguish authority to
explore from authority to modify, publish, deploy, or merge. Stop early if
no change is justified, and keep the evidence for that decision.
Template: templates/change-contract.md.
Discover before designing. Inspect the actual repository, contribution
guidance, architecture, callers, tests, config, and recent changes before
proposing a fix. Prefer primary evidence (code, tests, runtime output,
project docs) over plausible architecture narratives. Make unverified
assumptions and missing access explicit.
Select the smallest safe intervention. Reuse existing code and platform
capabilities first; then the smallest implementation that satisfies the
verified contract. Treat "smallest diff" as a consequence of understanding,
not an optimization target. Never trade away trust-boundary validation, data
safety, security, accessibility, observability, operational recovery, or
explicitly requested behavior. Record any deliberate ceiling and its upgrade
trigger.
Classify a test-hardening request explicitly when production behavior is
already correct and the work adds a focused guard for a coverage gap or named
mutation. That is a lightweight-path subtype, not an ordinary production
bugfix: use baseline-pass / mutant-fail evidence instead of requiring the new
test to fail on clean main. Load
references/lightweight-test-hardening.md
for the bounded contract and escalation triggers.
Execute by SDLC stage. Route the work to the stage that owns it —
discovery/requirements, design, implementation, verification, delivery,
learning. Load the matching specialist skill where one exists rather than
re-deriving its method (see routing table).
Keep an evidence ledger. Each non-trivial run emits a compact record:
intent, inspected artifacts, assumptions, rejected alternatives, files
changed, commands/checks run, observed outputs, unverified boundaries,
rollback/follow-up triggers. Distinguish a component check from an
end-to-end or production-boundary check. Format and rules:
references/evidence-ledger.md.
Template: templates/evidence-ledger.md.
Stop and escalate by rule. Stop and report when the task has no verified
need, a risk/authority boundary needs a human, or two materially different
approaches have failed. Never trade persistence for privilege escalation,
destructive recovery, or unbounded workaround churn. Rules:
references/risk-authority-gates.md.
For asynchronous work and stage-finality decisions, use
references/progress-and-finality.md:
judge semantic progress and declared dependencies, not elapsed time or
repeated observations.
Change-request work (conditional)
When the request is a change request — an issue, ticket, or tracked request that
will produce a pull request or an equivalent reviewable deliverable — the 6-step
core loop above still governs each stage, but the run also follows the canonical
9-phase change-request journey:
→ references/journey.md — intake → discovery →
design → spec → test plan → implementation → review → readiness → authorized
release.
The journey adds gates, a delivery packet for cross-phase state, and platform
lifecycle mechanics (GitHub or enterprise). It does not replace the core loop
or create a second orchestrator.
Do not load the journey for: a simple fix, refactor, or review that has no
issue/ticket trajectory. That work uses the core loop and the stage references
below without the journey's phases, packet, or gates.
Mid-flight engagement (conditional)
When the request is to continue delivery work that started elsewhere — a
half-implemented branch, an adopted spec, a change request another contributor
began — and no delivery packet exists, do not restart at phase 1 and do not
trust narrative summaries of what was done. Run position assessment first:
inventory observable artifacts, score them against phase exit conditions, emit
a position report (current step, satisfied gates with evidence pointers,
remaining checklist), then bootstrap a packet and continue under the normal
journey.
Procedure: references/position-assessment.md.
Load it only for mid-flight engagement; fresh change requests start at phase 1,
and existing packets govern resumability through
references/delivery-packet.md.
Path selection before ceremony
Select the delivery path immediately after current-state discovery and before
creating path-specific artifacts. Use the affected surface and risk, not line
count alone.
For a test-hardening request, choose the lightweight path only when production
behavior is already correct, production code remains unchanged, the change is
confined to tests or fixtures, and a focused deterministic check plus a bounded
controlled weakening can expose the regression. Record the invariant, the clean
baseline evidence, the named mutation, and the non-goals. Do not apply the
ordinary bugfix requirement that a new test fail on clean main.
For expensive or serialized CI, local design, hermeticity, focused tests,
targeted mutation, lint, compilation, and scope checks must pass before the
first push. Freeze the candidate before requesting one bounded final review or
the repository's required platform review. A timed-out review is inconclusive,
not a reason to launch repeated review rounds. Material changes invalidate the
relevant verification and bind the replacement verdicts to the new exact head.
The detailed test-hardening contract and escalation triggers are in
references/lightweight-test-hardening.md.
The one rule that defines "done"
"Done" is prohibited unless the declared verification target has actually been
exercised. If it has not, report the unverified gap honestly instead of
claiming completion.
A passing unit test is not the same as exercising the delivery boundary. A local
render is not production. State which boundary was checked and which was not.
Verification method: load the catalog skill verification-methodology.
Minimalism, correctly
Minimalism in this bundle is a conditional design choice made after
real-flow understanding — not an unconditional "fewest lines wins" reflex. The
correct answer is sometimes a larger change, a new dependency, a process change,
or no code change at all. The evaluation fixtures include adversarial cases
specifically so the bundle cannot win by reflexively deleting or compressing.
See references/stages.md §3.
Routing: compose, don't swallow
neckbeard owns the cross-stage contracts — the change contract, evidence
ledger, stop rules, and evaluation protocol. It does not own domain method.
When a stage has a specialist skill, load it and follow it. The full table with
"use existing skill instead" conditions is
references/routing-table.md. Summary:
| Stage / need |
Load this catalog skill instead of re-deriving |
| Stakeholder discovery, requirements, ACs |
product-discovery |
| User-facing behavior, interaction, information architecture |
product-design-and-ux |
| Formal specification, phase gates |
spec-driven-development |
| Reverse-engineering an existing codebase |
software-architecture-analysis |
| Designing or evolving an API / interface contract |
api-design-and-evolution |
| Root-cause debugging |
systematic-debugging |
| Security review, threat modeling, secure design |
secure-software-engineering |
| Accessibility (WCAG, keyboard/focus, error recovery) |
web-accessibility |
| Test strategy, regression testing, CI quality gates |
qa-methodology |
| Mobile app implementation (iOS/Android/Flutter/React Native) |
mobile-development |
| Docs / README / API reference |
technical-documentation |
| Verification verdicts and evidence |
verification-methodology |
| Intent contracts, work classification, autonomy gating, failure routing across a run |
bmad |
| Release planning, versioning, pipeline promotion, readiness, rollout, or rollback design |
release-engineering |
| Reliability, incident response, or operational recovery |
site-reliability-engineering |
If a specialist skill is not installed, neckbeard's stage references provide a
minimal fallback method — but note in the ledger that the specialist was absent.
Evaluation is a first-class deliverable
This bundle ships a versioned evaluation harness in eval/. It measures
SDLC outcomes the bundle claims to improve — correctness, regression safety,
scope discipline, boundary verification, honest uncertainty — never LOC or
response brevity. Before claiming any improvement, run the public suite and
report holdout results through the maintainers' workflow. Methodology:
references/evaluation.md.
Claims policy. Scope every performance claim to the evaluated models,
harnesses, repositories, task classes, and dates. Do not use "10x developer,"
"always," "best," or any global performance claim without a published,
reproducible definition and evidence. LOC may appear only as diagnostic
metadata, never as a success proxy.
File map
| Path |
Loaded when |
| references/stages.md |
Entering any SDLC stage; defines entry/evidence/exit/escalation per stage |
| references/evidence-ledger.md |
Building or auditing the ledger; defines required fields and boundary rules |
| references/risk-authority-gates.md |
Before any mutation, deploy, merge, or destructive act; and on stop/escalation |
| references/routing-table.md |
Deciding whether a specialist skill owns the current stage |
| references/journey.md |
Change-request work only — an issue/ticket/request that will produce a PR or equivalent reviewable deliverable. Defines the 9-phase intake→release sequence, gates, and paths. Not loaded for a simple fix, refactor, or review without an issue/ticket trajectory (that work uses the core loop above). |
| references/lightweight-test-hardening.md |
Test-only regression guards for already-correct production behavior: baseline-pass / mutant-fail evidence, hermeticity, finality-before-push, bounded review, and escalation triggers. |
| references/lifecycle.md |
Change-request work only — platform mechanics (GitHub reference mode or enterprise mode) for intake, submission, CI/review monitoring, and authorized release. Not loaded for a simple fix, refactor, or review without an issue/ticket trajectory. |
| references/delivery-packet.md |
Change-request work only — carrying state across phases of a change-request run, or resuming after a context boundary; defines packet fields, artifact ownership, lifecycle states, and resumability. Not loaded for a simple fix, refactor, or review without an issue/ticket trajectory. |
| references/position-assessment.md |
Mid-flight engagement only — continuing delivery work that started elsewhere with no delivery packet: artifact inventory, exit-condition scoring, position report, packet bootstrap. |
| references/tracker-discovery.md |
Change-request intake — detecting which tracking system holds the work item and routing tracker operations to the matching catalog tooling skill; keeps the journey vocabulary tracker-neutral. |
| references/evaluation.md |
Designing, running, or reporting an evaluation |
| templates/ |
Change contract, decision record, evidence ledger, verification plan, eval report; plus templates/delivery-packet.md — the fillable packet, for change-request work only (not for a simple fix/refactor/review without an issue trajectory) |
| eval/ |
Task schema, rubric, baseline protocol, fixtures, runner |
1---2name: neckbeard3description: >-4 Use when asked to fix, build, refactor, review, verify, or release software and the work5 is non-trivial — including delivering a change request through planning, gates,6 implementation, review, verified PR, and authorized release; continuing work started7 elsewhere with no packet via evidence-based position assessment; or operating tracked8 work items in Linear, Jira, or Notion via their tooling skills. neckbeard routes the9 change through framing, discovery, design, implementation, review, verification, and10 learning — choosing the smallest *safe* intervention, proving it at the real delivery11 boundary, and leaving an inspectable evidence ledger. For issue-to-PR work, conditionally12 loads a 9-phase journey with gates and lifecycle integration. Composes specialist catalog13 skills rather than replacing them. Not a persona or LOC-minimizer. Do not use this skill14 for unrelated requests; route to the nearest named specialist.15license: MIT16compatibility: Agent harness with file read/write, terminal, and skill loading. No network or runtime dependency required by the bundle itself.17metadata:18 source: https://skillmd.com/skills/magnus919/neckbeard19 spec-version: "1.0"20 tags: sdlc, delivery, evidence, verification, workflow21---2223# neckbeard2425A disciplined operating model for software delivery. It does not perform a26character. It makes an agent do six things reliably: frame the change, discover27the real system before designing, pick the smallest *safe* intervention, execute28by stage, record evidence, and stop or escalate when the evidence runs out.2930The name is a joke about the "10x developer in a Markdown file" trope. The31substance is the opposite of that trope: effectiveness here is earned by32observable outcomes and scoped claims, never asserted by a persona.3334## When to load this3536Load neckbeard when a request is a non-trivial software change — a bug to37diagnose, a feature to build, a refactor, a review, a release to verify — and38you need a bounded, stage-aware way to carry it to a defensible "done."3940## When not to use4142Do **not** load it for:43- A single factual question or lookup (answer directly).44- A one-line edit whose contract is already fully specified (just do it, but45 still verify at the boundary).46- A task already owned end-to-end by a more specific skill (route there; see47 [references/routing-table.md](references/routing-table.md)).4849## Core loop5051Every run moves through the same spine. Each stage has entry conditions,52required evidence, exit conditions, and escalation rules detailed in53[references/stages.md](references/stages.md).54551. **Frame the change contract.** State the user-visible problem, constraints,56 system boundary, risks, and explicit non-goals. Distinguish *authority to57 explore* from *authority to modify, publish, deploy, or merge*. Stop early if58 no change is justified, and keep the evidence for that decision.59 Template: [templates/change-contract.md](templates/change-contract.md).60612. **Discover before designing.** Inspect the actual repository, contribution62 guidance, architecture, callers, tests, config, and recent changes *before*63 proposing a fix. Prefer primary evidence (code, tests, runtime output,64 project docs) over plausible architecture narratives. Make unverified65 assumptions and missing access explicit.66673. **Select the smallest safe intervention.** Reuse existing code and platform68 capabilities first; then the smallest implementation that satisfies the69 verified contract. Treat "smallest diff" as a *consequence of understanding*,70 not an optimization target. Never trade away trust-boundary validation, data71 safety, security, accessibility, observability, operational recovery, or72 explicitly requested behavior. Record any deliberate ceiling and its upgrade73 trigger.7475 Classify a test-hardening request explicitly when production behavior is76 already correct and the work adds a focused guard for a coverage gap or named77 mutation. That is a lightweight-path subtype, not an ordinary production78 bugfix: use baseline-pass / mutant-fail evidence instead of requiring the new79 test to fail on clean `main`. Load80 [references/lightweight-test-hardening.md](references/lightweight-test-hardening.md)81 for the bounded contract and escalation triggers.82834. **Execute by SDLC stage.** Route the work to the stage that owns it —84 discovery/requirements, design, implementation, verification, delivery,85 learning. Load the matching specialist skill where one exists rather than86 re-deriving its method (see routing table).87885. **Keep an evidence ledger.** Each non-trivial run emits a compact record:89 intent, inspected artifacts, assumptions, rejected alternatives, files90 changed, commands/checks run, observed outputs, unverified boundaries,91 rollback/follow-up triggers. Distinguish a component check from an92 end-to-end or production-boundary check. Format and rules:93 [references/evidence-ledger.md](references/evidence-ledger.md).94 Template: [templates/evidence-ledger.md](templates/evidence-ledger.md).95966. **Stop and escalate by rule.** Stop and report when the task has no verified97 need, a risk/authority boundary needs a human, or two materially different98 approaches have failed. Never trade persistence for privilege escalation,99 destructive recovery, or unbounded workaround churn. Rules:100 [references/risk-authority-gates.md](references/risk-authority-gates.md).101102For asynchronous work and stage-finality decisions, use103[references/progress-and-finality.md](references/progress-and-finality.md):104judge semantic progress and declared dependencies, not elapsed time or105repeated observations.106107## Change-request work (conditional)108109When the request is a change request — an issue, ticket, or tracked request that110will produce a pull request or an equivalent reviewable deliverable — the 6-step111core loop above still governs each stage, but the run also follows the canonical1129-phase change-request journey:113114→ **[references/journey.md](references/journey.md)** — intake → discovery →115design → spec → test plan → implementation → review → readiness → authorized116release.117118The journey adds gates, a delivery packet for cross-phase state, and platform119lifecycle mechanics (GitHub or enterprise). It does **not** replace the core loop120or create a second orchestrator.121122**Do not load the journey for:** a simple fix, refactor, or review that has no123issue/ticket trajectory. That work uses the core loop and the stage references124below without the journey's phases, packet, or gates.125126### Mid-flight engagement (conditional)127128When the request is to **continue delivery work that started elsewhere** — a129half-implemented branch, an adopted spec, a change request another contributor130began — and no delivery packet exists, do not restart at phase 1 and do not131trust narrative summaries of what was done. Run position assessment first:132inventory observable artifacts, score them against phase exit conditions, emit133a position report (current step, satisfied gates with evidence pointers,134remaining checklist), then bootstrap a packet and continue under the normal135journey.136137Procedure: [references/position-assessment.md](references/position-assessment.md).138Load it only for mid-flight engagement; fresh change requests start at phase 1,139and existing packets govern resumability through140[references/delivery-packet.md](references/delivery-packet.md).141142## Path selection before ceremony143144Select the delivery path immediately after current-state discovery and before145creating path-specific artifacts. Use the affected surface and risk, not line146count alone.147148For a test-hardening request, choose the lightweight path only when production149behavior is already correct, production code remains unchanged, the change is150confined to tests or fixtures, and a focused deterministic check plus a bounded151controlled weakening can expose the regression. Record the invariant, the clean152baseline evidence, the named mutation, and the non-goals. Do not apply the153ordinary bugfix requirement that a new test fail on clean `main`.154155For expensive or serialized CI, local design, hermeticity, focused tests,156targeted mutation, lint, compilation, and scope checks must pass before the157first push. Freeze the candidate before requesting one bounded final review or158the repository's required platform review. A timed-out review is inconclusive,159not a reason to launch repeated review rounds. Material changes invalidate the160relevant verification and bind the replacement verdicts to the new exact head.161162The detailed test-hardening contract and escalation triggers are in163[references/lightweight-test-hardening.md](references/lightweight-test-hardening.md).164165## The one rule that defines "done"166167> "Done" is prohibited unless the declared verification target has actually been168> exercised. If it has not, report the unverified gap honestly instead of169> claiming completion.170171A passing unit test is not the same as exercising the delivery boundary. A local172render is not production. State which boundary was checked and which was not.173Verification method: load the catalog skill `verification-methodology`.174175## Minimalism, correctly176177Minimalism in this bundle is a *conditional* design choice made **after**178real-flow understanding — not an unconditional "fewest lines wins" reflex. The179correct answer is sometimes a larger change, a new dependency, a process change,180or no code change at all. The evaluation fixtures include adversarial cases181specifically so the bundle cannot win by reflexively deleting or compressing.182See [references/stages.md](references/stages.md) §3.183184## Routing: compose, don't swallow185186neckbeard owns the *cross-stage contracts* — the change contract, evidence187ledger, stop rules, and evaluation protocol. It does **not** own domain method.188When a stage has a specialist skill, load it and follow it. The full table with189"use existing skill instead" conditions is190[references/routing-table.md](references/routing-table.md). Summary:191192| Stage / need | Load this catalog skill instead of re-deriving |193|---|---|194| Stakeholder discovery, requirements, ACs | `product-discovery` |195| User-facing behavior, interaction, information architecture | `product-design-and-ux` |196| Formal specification, phase gates | `spec-driven-development` |197| Reverse-engineering an existing codebase | `software-architecture-analysis` |198| Designing or evolving an API / interface contract | `api-design-and-evolution` |199| Root-cause debugging | `systematic-debugging` |200| Security review, threat modeling, secure design | `secure-software-engineering` |201| Accessibility (WCAG, keyboard/focus, error recovery) | `web-accessibility` |202| Test strategy, regression testing, CI quality gates | `qa-methodology` |203| Mobile app implementation (iOS/Android/Flutter/React Native) | `mobile-development` |204| Docs / README / API reference | `technical-documentation` |205| Verification verdicts and evidence | `verification-methodology` |206| Intent contracts, work classification, autonomy gating, failure routing across a run | `bmad` |207| Release planning, versioning, pipeline promotion, readiness, rollout, or rollback design | `release-engineering` |208| Reliability, incident response, or operational recovery | `site-reliability-engineering` |209210If a specialist skill is not installed, neckbeard's stage references provide a211minimal fallback method — but note in the ledger that the specialist was absent.212213## Evaluation is a first-class deliverable214215This bundle ships a versioned evaluation harness in [eval/](eval/). It measures216SDLC *outcomes* the bundle claims to improve — correctness, regression safety,217scope discipline, boundary verification, honest uncertainty — never LOC or218response brevity. Before claiming any improvement, run the public suite and219report holdout results through the maintainers' workflow. Methodology:220[references/evaluation.md](references/evaluation.md).221222**Claims policy.** Scope every performance claim to the evaluated models,223harnesses, repositories, task classes, and dates. Do not use "10x developer,"224"always," "best," or any global performance claim without a published,225reproducible definition and evidence. LOC may appear only as diagnostic226metadata, never as a success proxy.227228## File map229230| Path | Loaded when |231|---|---|232| [references/stages.md](references/stages.md) | Entering any SDLC stage; defines entry/evidence/exit/escalation per stage |233| [references/evidence-ledger.md](references/evidence-ledger.md) | Building or auditing the ledger; defines required fields and boundary rules |234| [references/risk-authority-gates.md](references/risk-authority-gates.md) | Before any mutation, deploy, merge, or destructive act; and on stop/escalation |235| [references/routing-table.md](references/routing-table.md) | Deciding whether a specialist skill owns the current stage |236| [references/journey.md](references/journey.md) | **Change-request work only** — an issue/ticket/request that will produce a PR or equivalent reviewable deliverable. Defines the 9-phase intake→release sequence, gates, and paths. **Not loaded** for a simple fix, refactor, or review without an issue/ticket trajectory (that work uses the core loop above). |237| [references/lightweight-test-hardening.md](references/lightweight-test-hardening.md) | Test-only regression guards for already-correct production behavior: baseline-pass / mutant-fail evidence, hermeticity, finality-before-push, bounded review, and escalation triggers. |238| [references/lifecycle.md](references/lifecycle.md) | **Change-request work only** — platform mechanics (GitHub reference mode or enterprise mode) for intake, submission, CI/review monitoring, and authorized release. **Not loaded** for a simple fix, refactor, or review without an issue/ticket trajectory. |239| [references/delivery-packet.md](references/delivery-packet.md) | **Change-request work only** — carrying state across phases of a change-request run, or resuming after a context boundary; defines packet fields, artifact ownership, lifecycle states, and resumability. **Not loaded** for a simple fix, refactor, or review without an issue/ticket trajectory. |240| [references/position-assessment.md](references/position-assessment.md) | **Mid-flight engagement only** — continuing delivery work that started elsewhere with no delivery packet: artifact inventory, exit-condition scoring, position report, packet bootstrap. |241| [references/tracker-discovery.md](references/tracker-discovery.md) | **Change-request intake** — detecting which tracking system holds the work item and routing tracker operations to the matching catalog tooling skill; keeps the journey vocabulary tracker-neutral. |242| [references/evaluation.md](references/evaluation.md) | Designing, running, or reporting an evaluation |243| [templates/](templates/) | Change contract, decision record, evidence ledger, verification plan, eval report; plus [templates/delivery-packet.md](templates/delivery-packet.md) — the fillable packet, **for change-request work only** (not for a simple fix/refactor/review without an issue trajectory) |244| [eval/](eval/) | Task schema, rubric, baseline protocol, fixtures, runner |