Pp Agoda

<!-- GENERATED FILE — DO NOT EDIT.
     This file is a verbatim mirror of library/travel/agoda/SKILL.md,
     regenerated post-merge by tools/generate-skills/. Hand-edits here are
     silently overwritten on the next regen. Edit the library/ source instead.
     See the repository agent guide, section "Generated artifacts: registry.json, cli-skills/". -->

Agoda — Printing Press CLI

Prerequisites: Install the CLI

This skill drives the agoda-pp-cli binary. You must verify the CLI is installed before invoking any command from this skill. If it is missing, install it first:

1. Install via the Printing Press installer. It defaults binaries to $HOME/.local/bin on macOS/Linux and %LOCALAPPDATA%\Programs\PrintingPress\bin on Windows:

   npx -y @mvanhorn/printing-press-library install agoda --cli-only

2. Verify: agoda-pp-cli --version
3. Ensure the reported install directory is on $PATH for the agent/runtime that will invoke this skill.

If the npx install fails (no Node, offline, etc.), fall back to a direct Go install (requires Go 1.26.6 or newer). This installs into $GOPATH/bin (default $HOME/go/bin), so add that directory to $PATH instead:

go install github.com/mvanhorn/printing-press-library/library/travel/agoda/cmd/agoda-pp-cli@latest

If --version reports "command not found" after install, the runtime cannot see the binary directory on $PATH. Do not proceed with skill commands until verification succeeds.

Agoda returns both the advertised price and the true all-in price in the same response, but only ever shows you the advertised one. This CLI surfaces both, breaks out the hidden tax-and-fee delta, and re-sorts by real cost - which routinely changes which hotel is cheapest. It talks to Agoda over plain HTTP with no browser, no rendering service, and no API key, and it keeps a local price history so it can answer questions a stateless scraper cannot.

When to Use This CLI

Use this CLI when an agent needs real Agoda hotel prices, particularly when the question involves cost. It is the right tool for judging what a stay actually costs, comparing finalists, sweeping flexible dates for a price floor, and tracking price drops over time. Agoda's inventory is strongest in Asia-Pacific, so it is often the better source than a Booking.com or Google Hotels tool for destinations in that region.

Anti-triggers

Do not use this CLI for:

• Do not use this CLI to complete a booking or take payment - it surfaces prices and deep links, and the reservation itself happens on Agoda.
• Do not use it for flights, activities, airport transfers, or car rental; it is scoped to hotels only.
• Do not use it to compare Agoda against other travel sites - it reads Agoda only, so a cross-OTA question needs a different tool.
• Do not use it to cancel, modify, or refund an existing reservation.

Unique Capabilities

These capabilities aren't available in any other tool for this API.

Honest pricing

• hotels search — Shows what you will actually pay, not the teaser rate, with the hidden tax-and-fee delta broken out per property.

  Reach for this instead of any scraped Agoda price. Quote the inclusive figure to a user; the advertised rate is not what they will be charged.

  agoda-pp-cli hotels search Tokyo --checkin 2026-10-15 --nights 2 --adults 2 --currency USD --agent

• hotels rank — Re-sorts a destination's results by true all-in price instead of Agoda's teaser-price ranking.

  Use this whenever the decision is about price. Ordinary search ordering inherits Agoda's advertised-price ranking and will mislead.

  agoda-pp-cli hotels rank Tokyo --checkin 2026-10-15 --nights 2 --limit 10 --agent

• hotels fees — Flags properties whose tax-and-fee ratio is an outlier against the destination median.

  Use before recommending a property. A hotel with a below-median advertised price and an above-median fee ratio is the classic bait pattern.

  agoda-pp-cli hotels fees Tokyo --checkin 2026-10-15 --nights 2 --agent

Local state that compounds

• prices cheapest — Returns the cheapest check-in dates across a flexible window for a destination.

  Use for flexible-date travelers. Returns the price floor across a window rather than a single-date quote.

  agoda-pp-cli prices cheapest Tokyo --window 2026-10-01..2026-11-30 --nights 3 --agent

• vip delta — Runs the same search signed-in and anonymous, then diffs per property to show what your VIP tier is actually worth.

  Use when a user asks whether signing in or chasing a VIP tier is worth it. Reports the measured discount on a real search instead of marketing copy.

  agoda-pp-cli vip delta Tokyo --checkin 2026-10-15 --nights 2 --agent

• watch run — Surfaces only watched properties whose latest true all-in price dropped meaningfully below their trailing median.

  Schedule it. Returns empty most days and returns something worth acting on when a watched property actually drops.

  agoda-pp-cli watch run --min-pct 7 --agent

• search — Full-text search over every property this CLI has already seen, with no network call.

  Use after a few live searches to answer property questions without spending a request or waiting on the network.

  agoda-pp-cli search "shinjuku" --agent

Agent-native plumbing

• compare — Puts finalist properties side by side on true all-in price, hidden fee share, review score, star rating, and free-cancellation deadline.

  Use once the choice is narrowed to finalists, instead of re-reading two detail pages and eyeballing the difference.

  agoda-pp-cli compare 936623 788273 --destination Tokyo --checkin 2026-10-15 --nights 2 --agent

Command Reference

destinations — Resolve a destination name to the numeric city id every Agoda search requires

• agoda-pp-cli destinations — Resolve a free-text destination (city, area, landmark) to an Agoda city id

reviews — Guest reviews for a property, paginated and sortable

• agoda-pp-cli reviews — List guest reviews for a property by Agoda hotel id

Finding the right command

When you know what you want to do but not which command does it, ask the CLI directly:

agoda-pp-cli which "<capability in your own words>"

which resolves a natural-language capability query to the best matching command from this CLI's curated feature index. Exit code 0 means at least one match; exit code 2 means no confident match — fall back to --help or use a narrower query.

Recipes

What will this actually cost

agoda-pp-cli hotels search Tokyo --checkin 2026-10-15 --nights 2 --adults 2 --currency USD --agent --select results.name,results.price_all_in,results.price_advertised,results.hidden_pct

Returns just the four fields that matter for a cost decision, keeping the deeply nested Agoda payload out of the agent's context.

The cheapest hotel is not the one listed cheapest

agoda-pp-cli hotels rank Tokyo --checkin 2026-10-15 --nights 2 --limit 10 --agent

Re-sorts by all-in cost; properties with above-average fee loads drop down the list and genuinely cheaper stays surface.

Find the price floor for a flexible trip

agoda-pp-cli prices cheapest Tokyo --window 2026-10-01..2026-11-30 --nights 3 --agent

Sweeps a two-month window in one pass and returns the cheapest check-in dates rather than a single-date quote.

Spot the resort-fee trap

agoda-pp-cli hotels fees Tokyo --checkin 2026-10-15 --nights 2 --agent

Ranks properties by how much of their true cost is tax and fees, flagging outliers against the destination median.

Is signing in worth anything here

agoda-pp-cli vip delta Tokyo --checkin 2026-10-15 --nights 2 --agent

Issues the same search authenticated and anonymous and reports the measured per-property discount.

Auth Setup

Public hotel search, destination lookup, property detail, and reviews need no credentials at all - they replay over ordinary HTTP. Only member-priced and account surfaces (saved properties, AgodaVIP tier, vip delta) need a logged-in session: copy the Cookie header from a signed-in agoda.com browser tab and export it as AGODA_COOKIE (AGODA_SESSION_COOKIE is also accepted), and subsequent authenticated calls replay with it.

Run agoda-pp-cli doctor to verify setup.

Agent Mode

Add --agent to any command. Expands to: --json --compact --no-input --no-color.

• Pipeable — JSON on stdout, errors on stderr

• Filterable — --select keeps a subset of fields. Dotted paths descend into nested structures; arrays traverse element-wise. Critical for keeping context small on verbose APIs:

  agoda-pp-cli reviews --agent --select hotelReviewId,rating,reviewComments

• Previewable — --dry-run shows the request without sending

• Non-interactive — never prompts, every input is a flag

• Read-only — do not use this CLI for create, update, delete, publish, comment, upvote, invite, order, send, or other mutating requests

Paths and state

Agents should treat the CLI's path resolver as part of the runtime contract:

• Use --home <dir> for one invocation, or set AGODA_HOME=<dir> to relocate all four path kinds under one root.

• Use per-kind env vars only when a specific kind must diverge: AGODA_CONFIG_DIR, AGODA_DATA_DIR, AGODA_STATE_DIR, AGODA_CACHE_DIR.

• Resolution order is per-kind env var, --home, AGODA_HOME, XDG (XDG_CONFIG_HOME, XDG_DATA_HOME, XDG_STATE_HOME, XDG_CACHE_HOME), then platform defaults.

• config contains settings like config.toml and profiles. data contains credentials.toml, data.db, cookies, and auth sidecars. state contains persisted queries, jobs, and teach.log. cache contains regenerable HTTP/cache files.

• Stored secrets live in credentials.toml under the data dir. Existing legacy config.toml secrets are read for compatibility and leave config.toml on the first auth write.

• Run agoda-pp-cli doctor --fail-on warn to surface path and credential-location warnings. agent-context exposes a schema v4 paths block for agents that need the resolved dirs.

• For MCP, pass relocation through the MCP host config. The MCP binary does not inherit CLI flags:

  {
    "mcpServers": {
      "agoda": {
        "command": "agoda-pp-mcp",
        "env": {
          "AGODA_HOME": "/srv/agoda"
        }
      }
    }
  }

Fleet precedence: an inherited per-kind env var overrides an explicit --home for that kind. Use AGODA_HOME or per-kind vars as durable fleet levers, and use --home only for a single invocation. Relocation is not reversible by unsetting env vars; move files manually before clearing AGODA_HOME, or doctor will not find credentials left under the former root.

Automatic learning

This CLI ships a self-capturing learning loop. The CLI does its own bookkeeping: every invocation is journaled locally, a failed flag followed by a corrected retry auto-derives a flag_alias candidate, and a teach on a query family without a playbook auto-synthesizes a playbook_candidate from the session's journal. Your job is judgment only: recall first, act on surfaced candidates, teach the final answer, playbook amend when you observe a correction. You never record failures by hand.

Step 1: recall before any discovery

Before list/search/drill commands on a new user question, run:

agoda-pp-cli recall "<user's question>" --agent

The response envelope:

{
  "query": "...",
  "normalized": "<normalized form>",
  "query_entities": ["..."],
  "found": true | false,
  "match_score": 0.0,
  "results": [
    { "resource_id": "...", "resource_type": "...", "venue": "...",
      "confidence": 2, "entity_match": "exact|partial|unknown",
      "source": "taught|preseed|pattern", "warnings": ["..."] }
  ],
  "mismatches": [ /* only when --debug-mismatches */ ],
  "warnings": [ /* top-level */ ],
  "candidates": [
    { "id": 12, "class": "flag_alias | playbook_candidate",
      "summary": "...", "sightings": 3, "last_seen": "...",
      "rationale": "...",
      "next_action": ["<trial command>", "agoda-pp-cli learnings confirm 12"] }
  ],
  "playbook": {
    "query_family": "...",
    "playbook": {
      "steps": [ { "cmd": "<command with {slot} substitution>", "purpose": "..." } ],
      "entity_slots": ["$ENTITY"],
      "expected_tool_calls": 3
    },
    "slots_resolved": { "$ENTITY": { "token": "<live token>", "canonical": "<canonical>" } },
    "notes": "<workarounds + gotchas for this query family>"
  },
  "notes": "<duplicate surface for non-playbook callers>"
}

Empty-store short-circuit: if the store has no learnings, playbooks, or candidates yet (recall finds nothing and learnings list and learnings candidates are both empty), skip recall for the rest of this session instead of taxing every query; resume recall-first once something has been taught.

Step 2: decision tree

Read candidates, playbook, notes, results[0], and warnings in that order:

if Candidates present (warnings include "candidates_present"):
    -> candidates are try-then-confirm, never facts. Follow each candidate's
       two-step next_action verbatim: run the trial command first, then run
       `learnings confirm <id>` only after the trial verified the behavior.
       Reject a wrong candidate with `learnings reject <id>`.
    -> NEVER re-teach something recall surfaced as a candidate; confirm or
       reject that candidate instead of teaching a duplicate.
    -> candidates ride alongside playbooks and resource hits, not instead of
       them; continue with the branches below after acting on them.

if Playbook present:
    -> READ Playbook.notes verbatim FIRST (workarounds + gotchas the CLI surface doesn't expose)
    -> replay Playbook.steps in order, substituting Playbook.slots_resolved entries
       for the entity slot tokens. If a step's slot is unresolved, fall back to
       discovery for that step only.
    -> the Playbook's expected_tool_calls is a budget; if you find yourself running
       materially more, record the divergence via `agoda-pp-cli playbook amend`
       at end-of-session.

elif Notes present (no Playbook):
    -> read Notes verbatim before any discovery step; they carry known gotchas
       for this query family even when no structured choreography exists yet.

elif Found AND Results[0].EntityMatch == "exact" AND Results[0].Confidence >= 2:
    -> skip discovery; fetch live data for Results[*].ResourceID in parallel

elif Found AND Results[0].EntityMatch == "partial":
    -> candidate hint, NOT a hit; read the resource title to validate before trusting

elif (any row in Mismatches[] when --debug-mismatches was passed):
    -> treat as cold start; the stored learning is for a different entity
       (different canonical resolved from query_entities)

else:  // Found == false, no playbook, no notes
    -> cold start; run discovery normally; teach the answer afterward (Step 4).
       If the family has no playbook yet, that teach auto-synthesizes a
       playbook candidate from this session's journal - you do not need to
       record one by hand.

Playbook and Notes are orthogonal to the per-resource path. A recall response can carry both a Playbook AND a Results[] hit - use both: the Playbook tells you which choreography to run; the resource hits short-circuit specific steps. Default to skipping mismatches; pass --debug-mismatches only when investigating cold-start surprises.

Candidate judgment details: learnings confirm <id> prints the candidate's full payload before materializing it - check that the printed payload matches the behavior you verified. learnings reject <id> tombstones the derivation signature so the same candidate does not resurface. The envelope carries only the few candidates worth acting on now; agoda-pp-cli learnings candidates lists the full open set.

Graceful degradation: if learnings confirm is an unknown command, you are driving an older binary - ignore the candidates guidance and follow the rest of the protocol.

Step 3: always read warnings

• low_confidence: row exists at confidence<2. Treat as a hint, not a skip-discovery hit.
• resource_not_in_store: the local store doesn't have the resource the learning points at. The match validator couldn't classify entities — direct-fetch and re-evaluate.
• cross_alias_match (per-result): the row was taught under a different alias and matched the live query's canonical via entity_lookups (e.g., a "USA" teach satisfying a "United States" recall). Trust the resource_id.
• similar_shape_different_entity:<canonical> (top-level): a structurally matching row exists but its canonical entity differs from the live query's. Treated as cold start; the warning carries the conflicting canonical as a hint, but the row is NOT promoted into Results.
• ambiguous_alias (top-level): a single query entity resolved to multiple canonicals (e.g., "Cards" → Arizona Cardinals + St. Louis Cardinals). Surface the ambiguity from context before committing to a resource.
• candidates_present (top-level): the envelope carries a candidates section. Handle it via the candidates branch in Step 2 before anything else.
• Top-level no_learnings_for_query_family: the table had no rows above the Jaccard floor. Pure cold start.