# AGENTS.md — Build guide for Debriefing

## Project scope
Fetch selected public competitor pages on a bounded schedule, compare meaningful text changes and create a reviewed digest. Let the user exclude repeated headers and volatile regions before optional AI summarization.

Catalogue verdict: kinda. Diffing a competitor's pricing page on a cron and asking a model what changed is genuinely an afternoon. changedetection.io will do the watching for you before you write a line. The gap is everything between a diff and a brief. Most page changes are noise: a rotated testimonial, a reordered nav, a CDN hash. Deciding which changes are real, tying them to hiring and funding signals, and turning that into three sentences a founder acts on is judgement encoded over many iterations. Build it and your first month is mostly you deleting alerts about nothing.
Use the implementation prompt below to define the deliverable. Complete each phase's acceptance checks before extending the scope.

## Working agreement
- Inspect the repository and its existing instructions before choosing paths, dependencies or commands. Keep one coherent stack and explain changes to the proposed architecture.
- Plan a vertical slice that accepts a real input and produces the useful output described below. Persist only the state the prompt calls for; respect memory-only and upstream-managed workflows. Use fixtures only when they are clearly labelled.
- After scaffolding, document the actual install, development, check and build commands in README and keep them synchronized with the package or project manifest. Do not report commands as successful unless they ran.
- Work in small steps. At handoff, list implemented flows, checks actually performed, remaining blockers, and any credentials or provider setup the owner must supply.
- Do not publish, spend money, contact customers, delete source data or run irreversible migrations without the project owner's authorization.

## Prerequisites
- A Python virtual environment, writable input/output directories and sufficient disk for both originals and outputs. Bind the service to localhost.
- Implementation components: Python, FastAPI and server-rendered HTML with HTMX for a local interface. SQLite for manifests and job state, with an explicit worker process and immutable source files. HTTP crawling with HTML parsing and an optional bounded Playwright renderer for user-authorized sites.
- Scope boundary: noise filtering, which is the actual product: most diffs are rotated testimonials and changed asset hashes, not competitive moves; horizon watch, meaning the substitutes and new entrants you did not think to add to the list

## Stack and architecture
- Python, FastAPI and server-rendered HTML with HTMX for a local interface.
- SQLite for manifests and job state, with an explicit worker process and immutable source files.
- HTTP crawling with HTML parsing and an optional bounded Playwright renderer for user-authorized sites.
- Domain model: competitors, approved URLs, fetch snapshots, normalized text, diff exclusions and source-linked briefs

## Security and data integrity
- Bound file sizes and processing time, reject path traversal, and use argument arrays for subprocesses. Treat imported text as data and redact confidential source content from logs. Constrain crawl hosts, block private-network URLs and redirect pivots, and respect crawl delays and access restrictions.
- Correctness boundary: Never replace a prior snapshot with a failed fetch; summaries distinguish observed changes from speculative business implications.
- Save a job manifest with input hash, parameters and state. Write to temporary outputs, then atomically finalize only successful results; resume unfinished jobs without replacing originals.
- Export sources, manifests and outputs with checksums. Keep failed-job diagnostics and allow retry into a new output path; restore the database and file directory together.

## Agent implementation rules
- Project rule — data model: competitors, approved URLs, fetch snapshots, normalized text, diff exclusions and source-linked briefs
- Project rule — preserve this invariant: Never replace a prior snapshot with a failed fetch; summaries distinguish observed changes from speculative business implications.
- Project rule — acceptance evidence: A rotated timestamp produces no alert after an exclusion rule; a changed pricing paragraph links to both dated snapshots and the exact diff.

## Optional agent skills and references
- Optional external skill: [seo-audit](https://github.com/coreyhaines31/marketingskills/blob/main/skills/seo-audit/SKILL.md) — Investigate crawlability, indexing, page metadata, internal linking and on-page content issues. Review its instructions and compatibility before use; it does not grant deployment, data-access or publication permission.
- Optional external skill: [modern-python](https://github.com/trailofbits/skills/blob/main/plugins/modern-python/skills/modern-python/SKILL.md) — Set up Python projects with pyproject.toml, dependency management, linting, typing and automated checks. Review its instructions and compatibility before use; it does not grant deployment, data-access or publication permission.
- Optional external skill: [agent-browser](https://github.com/vercel-labs/agent-browser/blob/main/skills/agent-browser/SKILL.md) — Automate browser interaction using accessibility snapshots, element references and reproducible navigation workflows. Review its instructions and compatibility before use; it does not grant deployment, data-access or publication permission.

Read the linked SKILL.md and its dependencies before adding a skill. Select only the skills matching this project's runtime and task; their documentation does not supply API access, credentials or approval to perform external actions. Pin the reviewed revision where the tool supports it. Follow the chosen agent's documented project-level installation mechanism.

## Distribution ideas
These are optional planning notes. Obtain the owner's approval before publishing or contacting anyone.
- Demonstrate the actual Debriefing-inspired workflow with owned or clearly labeled sample data: Fetch selected public competitor pages on a bounded schedule, compare meaningful text changes and create a reviewed digest. Let the user exclude repeated headers and volatile regions before optional AI summarization.
- Publish a reproducible walkthrough with this observable result: A rotated timestamp produces no alert after an exclusion rule; a changed pricing paragraph links to both dated snapshots and the exact diff.
- Explain who can operate this scoped tool, its setup and ongoing costs, and these remaining product gaps: noise filtering, which is the actual product: most diffs are rotated testimonials and changed asset hashes, not competitive moves; horizon watch, meaning the substitutes and new entrants you did not think to add to the list Avoid guaranteed savings, performance scores or implied endorsement.

## Engineering roadmap
1. Phase 1 — Scope and fixtures. Implement this bounded workflow: Fetch selected public competitor pages on a bounded schedule, compare meaningful text changes and create a reviewed digest. Let the user exclude repeated headers and volatile regions before optional AI summarization. Record prerequisites, select representative user-owned fixtures and document the unsupported features: noise filtering, which is the actual product: most diffs are rotated testimonials and changed asset hashes, not competitive moves; horizon watch, meaning the substitutes and new entrants you did not think to add to the list
2. Phase 2 — Durable model. Model competitors, approved URLs, fetch snapshots, normalized text, diff exclusions and source-linked briefs Add migrations or a versioned document format, explicit validation, stable IDs and a visible import-error report. Preserve this rule: Never replace a prior snapshot with a failed fetch; summaries distinguish observed changes from speculative business implications.
3. Phase 3 — Complete the first useful path. Implement the workflow's input, review and output interface, with clear controls and explicit empty/error states. Save a job manifest with input hash, parameters and state. Write to temporary outputs, then atomically finalize only successful results; resume unfinished jobs without replacing originals.
4. Phase 4 — Permissions and integration failure. Bound file sizes and processing time, reject path traversal, and use argument arrays for subprocesses. Treat imported text as data and redact confidential source content from logs. Constrain crawl hosts, block private-network URLs and redirect pivots, and respect crawl delays and access restrictions. Request integration credentials and permissions only for the enabled feature; show a disconnected state instead of mock results.
5. Phase 5 — Portable handoff. Export sources, manifests and outputs with checksums. Keep failed-job diagnostics and allow retry into a new output path; restore the database and file directory together. Include setup, operating limits, fixture walkthrough and shutdown/restart instructions in the README.
6. Phase 6 — Acceptance scenarios. A rotated timestamp produces no alert after an exclusion rule; a changed pricing paragraph links to both dated snapshots and the exact diff. Repeat the workflow after restart and with a denied permission or unavailable dependency; show recoverable failure rather than a success placeholder.

## Paid-product capabilities outside this build
- noise filtering, which is the actual product: most diffs are rotated testimonials and changed asset hashes, not competitive moves
- horizon watch, meaning the substitutes and new entrants you did not think to add to the list
- the non-page signals stitched into the same brief: hiring, funding, traffic and LinkedIn movement
- the analytical step from what changed to why it matters to what to do next
- an archive going back far enough that a change reads as a trend rather than an event

## Implementation prompt
WORKING SLICE
Fetch selected public competitor pages on a bounded schedule, compare meaningful text changes and create a reviewed digest. Let the user exclude repeated headers and volatile regions before optional AI summarization.

Build this scoped Debriefing-inspired workflow with a documented data model and visible failure states.

Architecture
- Python, FastAPI and server-rendered HTML with HTMX for a local interface.
- SQLite for manifests and job state, with an explicit worker process and immutable source files.
- HTTP crawling with HTML parsing and an optional bounded Playwright renderer for user-authorized sites.

Prerequisites and limits
A Python virtual environment, writable input/output directories and sufficient disk for both originals and outputs. Bind the service to localhost.
Outside this release: noise filtering, which is the actual product: most diffs are rotated testimonials and changed asset hashes, not competitive moves; horizon watch, meaning the substitutes and new entrants you did not think to add to the list

Data model and correctness
competitors, approved URLs, fetch snapshots, normalized text, diff exclusions and source-linked briefs
Invariant: Never replace a prior snapshot with a failed fetch; summaries distinguish observed changes from speculative business implications.
Save a job manifest with input hash, parameters and state. Write to temporary outputs, then atomically finalize only successful results; resume unfinished jobs without replacing originals.

Security and privacy
Bound file sizes and processing time, reject path traversal, and use argument arrays for subprocesses. Treat imported text as data and redact confidential source content from logs. Constrain crawl hosts, block private-network URLs and redirect pivots, and respect crawl delays and access restrictions.

Recovery and export
Export sources, manifests and outputs with checksums. Keep failed-job diagnostics and allow retry into a new output path; restore the database and file directory together.

Implementation order
1. Phase 1 — Scope and fixtures. Implement this bounded workflow: Fetch selected public competitor pages on a bounded schedule, compare meaningful text changes and create a reviewed digest. Let the user exclude repeated headers and volatile regions before optional AI summarization. Record prerequisites, select representative user-owned fixtures and document the unsupported features: noise filtering, which is the actual product: most diffs are rotated testimonials and changed asset hashes, not competitive moves; horizon watch, meaning the substitutes and new entrants you did not think to add to the list
2. Phase 2 — Durable model. Model competitors, approved URLs, fetch snapshots, normalized text, diff exclusions and source-linked briefs Add migrations or a versioned document format, explicit validation, stable IDs and a visible import-error report. Preserve this rule: Never replace a prior snapshot with a failed fetch; summaries distinguish observed changes from speculative business implications.
3. Phase 3 — Complete the first useful path. Implement the workflow's input, review and output interface, with clear controls and explicit empty/error states. Save a job manifest with input hash, parameters and state. Write to temporary outputs, then atomically finalize only successful results; resume unfinished jobs without replacing originals.
4. Phase 4 — Permissions and integration failure. Bound file sizes and processing time, reject path traversal, and use argument arrays for subprocesses. Treat imported text as data and redact confidential source content from logs. Constrain crawl hosts, block private-network URLs and redirect pivots, and respect crawl delays and access restrictions. Request integration credentials and permissions only for the enabled feature; show a disconnected state instead of mock results.
5. Phase 5 — Portable handoff. Export sources, manifests and outputs with checksums. Keep failed-job diagnostics and allow retry into a new output path; restore the database and file directory together. Include setup, operating limits, fixture walkthrough and shutdown/restart instructions in the README.
6. Phase 6 — Acceptance scenarios. A rotated timestamp produces no alert after an exclusion rule; a changed pricing paragraph links to both dated snapshots and the exact diff. Repeat the workflow after restart and with a denied permission or unavailable dependency; show recoverable failure rather than a success placeholder.

Acceptance
A rotated timestamp produces no alert after an exclusion rule; a changed pricing paragraph links to both dated snapshots and the exact diff.
Use real source data or clearly labeled fixtures. Explain unsupported input and provider failures; do not fabricate analytics, delivery receipts, accuracy claims or security guarantees.

Optional agent guidance
Optional external skill: [seo-audit](https://github.com/coreyhaines31/marketingskills/blob/main/skills/seo-audit/SKILL.md) — Investigate crawlability, indexing, page metadata, internal linking and on-page content issues. Review its instructions and compatibility before use; it does not grant deployment, data-access or publication permission.
Optional external skill: [modern-python](https://github.com/trailofbits/skills/blob/main/plugins/modern-python/skills/modern-python/SKILL.md) — Set up Python projects with pyproject.toml, dependency management, linting, typing and automated checks. Review its instructions and compatibility before use; it does not grant deployment, data-access or publication permission.
Optional external skill: [agent-browser](https://github.com/vercel-labs/agent-browser/blob/main/skills/agent-browser/SKILL.md) — Automate browser interaction using accessibility snapshots, element references and reproducible navigation workflows. Review its instructions and compatibility before use; it does not grant deployment, data-access or publication permission.
Project rule — data model: competitors, approved URLs, fetch snapshots, normalized text, diff exclusions and source-linked briefs
Project rule — preserve this invariant: Never replace a prior snapshot with a failed fetch; summaries distinguish observed changes from speculative business implications.
Project rule — acceptance evidence: A rotated timestamp produces no alert after an exclusion rule; a changed pricing paragraph links to both dated snapshots and the exact diff.

## Completion evidence
Demonstrate the prompt's acceptance scenarios against the scoped workflow. Include setup from a clean checkout and failure recovery. Check persistence across restart and export/restore only for the state the prompt says to store; for memory-only tools, confirm that temporary content is discarded as specified. Record actual results and remaining limitations. A detailed plan alone does not establish a working replacement.