Skip to main content
Glama
cloudlens

cloudlens

cloudlens

Cloud cost findings priced at your negotiated rates, reconciled to your actual invoice, and delivered as changes you can merge.

Runs on your machine. Reads your bill and your code. Sends nothing anywhere.

cloudlens check --profile your-aws-profile
About $210.44 a month.
$141.26 a month looks avoidable (67% of the bill).

$137.39/month — Client VPN endpoint billed continuously with nobody connecting
  Why: 4,656 endpoint-hours billed, with 0.0 connection-hours
  Do:  Delete the endpoint, or disassociate its subnets.
  Risk if wrong: A VPN kept for break-glass access is unused by design.

$3.23/month — Amazon DynamoDB is just over its free allowance
  Why: 24,056 unit-hours billed against a free allowance of 18,600
  Do:  Reduce to 18,600 or below and the charge disappears entirely.

That is the whole product. Everything below is optional depth.

Or ask an agent instead of running a command — see Use it from Claude.


Why another one

Three things this does that the others do not.

It knows what you actually pay. The AWS Price List API is documented as "for informational purposes only". Every tool that prices findings from it — including the pre-deploy estimators — is wrong for any customer with an enterprise discount, commonly by 20–40%, always in the direction of overstating. cloudlens derives your real rate card from your own bill:

Rate card  4 rates derived from your own billing data (4 measured)
Observed discount off public pricing: 16.0% — pricing findings at list price
would overstate savings by roughly this much.

It refuses to publish numbers that don't tie out. Every run reconciles computed spend against your real invoice. If it doesn't match, you get the discrepancy instead of a savings figure.

┏━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━┳━━━━━━━━┳━━━━━━━━━━┓
┃ Period  ┃ Computed ┃ Invoice ┃  Delta ┃          ┃
┡━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━╇━━━━━━━━╇━━━━━━━━━━┩
│ 2026-07 │  $253.46 │ $253.46 │ +0.00% │ ties out │
└─────────┴──────────┴─────────┴────────┴──────────┘

It proves savings against history rather than projecting them.

"Had this been applied at the start of 2026-06, your actual invoices for 2026-06 and 2026-07 would have been $27.22 and $27.22 lower — computed against your own line items."


Related MCP server: AWS FinOps MCP Server

Install

Nothing to install:

uvx --from cloudlens-mcp cloudlens check --profile prod

(--from names the package, cloudlens names the command inside it. Bare uvx cloudlens-mcp starts the MCP server instead, which is the other half.)

Or clone it, to run the tests or work on it:

git clone <repo> && cd cloudlens
uv sync --extra dev        # or: pip install -e ".[dev]"
uv run cloudlens demo
uv run pytest

Python 3.11+. Six dependencies: duckdb, boto3, rich, typer, openpyxl, mcp. The lockfile is committed, so installs are reproducible.


Use

Without a repo — works on day one

Source access is a hard ask on a first engagement, so everything here comes from the bill alone. No CloudWatch, no agent, no describe permission.

cloudlens scan     --cur s3://bucket/prefix/ --invoices invoices.json
cloudlens trends   --cur s3://bucket/prefix/
cloudlens profile  --cur s3://bucket/prefix/

An hourly CUR is a time series, not a ledger — so the weekly rhythm of every resource is already in it:

i-0uatworker   billed 1,464h · active 405h · 72% idle · $55.34

        00              06              12              18
  Sun  ························
  Mon  ·········█████████······
  Fri  ·········█████████······
  Sat  ························

With a repo — ownership, AI defects, mergeable fixes

cloudlens findings --cur s3://... --repo . --invoices invoices.json --detail
cloudlens report   --cur s3://... --repo . --out report.xlsx
cloudlens fix      --cur s3://... --repo . --id gp2-to-gp3-001 --write

Attribution resolves ownership through a ten-rung ladder — code first, because git history outlives CloudTrail's 90-day window:

Attribution  92.4% of $1,091.56 traced to an owner
│    1 │ infrastructure as code          │ $433.84 │ 53.2% │
│    3 │ AWS-generated tag               │ $302.40 │ 37.1% │
│    — │ no owner found                  │  $38.64 │  4.7% │

In CI

cloudlens ci --cur s3://... --plan tfplan.json --repo . \
             --max-monthly-increase 500

Posts a PR comment nothing else can produce — the cost of the change at your rates, plus existing waste attached to the code being touched.


Commands

doctor

Run this first on someone else's account. What the credential can and cannot see

sweep

Every account you can reach, ranked by what is avoidable

snapshot

Record today, so a later check can say whether the bill actually moved

check

Start here. What is wasting money, in plain English

scan

Rate card and reconciliation

trends · profile

What changed, what spiked, weekly usage rhythm

attribute · ask · answer

Who owns what; ask a human; record the answer

findings

Waste, priced and citing AWS best practices

aiscan

LLM cost defects in source — unbounded loops, uncapped prompts

fix · tagplan

Findings as mergeable diffs

ci

Price a Terraform plan, comment on the PR

report

Six-sheet spreadsheet for finance


Giving it to someone else

docs/FIRST-RUN.md is the page to send them — four commands from an empty laptop, including installing uv and pointing it at an account.

Run this first, on any account that is not your own:

uvx --from cloudlens-mcp cloudlens doctor --profile theirs

It checks the credential against every call the tool makes and says what each permission buys. This matters because cloudlens degrades quietly — a missing permission makes one analyser return nothing, which reads exactly like that analyser finding nothing, so a half-blind report looks like a clean account.

docs/iam-readonly-role.json grants exactly the 18 actions it needs and denies everything else by NotAction, so the grant and the guarantee are the same list. Every action is a read; there is no mutating AWS call anywhere in the codebase.

The minimum to get any answer at all is ce:GetCostAndUsage plus cloudwatch:ListMetrics and cloudwatch:GetMetricStatistics. Everything else adds findings rather than enabling the tool.

Use it from Claude

Two ways, both read-only.

MCP server. Copy .mcp.json.example to .mcp.json:

{
  "mcpServers": {
    "cloudlens": {
      "command": "uvx",
      "args": ["cloudlens-mcp@latest"],
      "env": { "AWS_PROFILE": "your-readonly-profile" }
    }
  }
}

Nothing to clone and nothing to install. uvx fetches the package, builds a throwaway environment and runs it. The CLI is the same package, asked for by command name:

uvx --from cloudlens-mcp cloudlens check --profile prod

Replace @latest with ==0.8.0 when you want a pinned version.

Six tools: what_is_wasting_money, what_was_not_examined, what_changed, free_tier_status, show_fix, capabilities.

Skill. One command, which also has nothing to clone:

uvx --from cloudlens-mcp cloudlens install

It writes the skill to ~/.claude/skills/cloudlens/ and prints the MCP entry to paste. --skills-dir puts it elsewhere, --mcp-config .mcp.json writes the entry rather than printing it.

The skill teaches an agent how to read the output — which confidence tiers can be added together, why a short findings list needs the coverage number beside it, and always to relay the failure case before recommending a deletion. The MCP server carries the essentials in its own instructions, which arrive over the wire, so the two agree by construction: the skill ships inside the package next to the server that describes it.

On guardrails

AWS's Agent Toolkit needs aws:CalledViaAWSMCP deny policies so an agent cannot do more through MCP than the human should, because that toolkit can call anything.

This needs none. There is no mutating AWS call anywhere in the codebase — the tool names contain no verbs, show_fix returns a diff for you to review rather than applying it, and the only file ever written goes into your own working copy. There is no privilege to escalate, so there is nothing to fence off. Ask the capabilities tool and it will say so itself.

The credential is the AWS profile already on the machine. Nothing new is issued, nothing is stored, and nothing leaves the host.


Access

Minimum is read on one S3 prefix. See docs/SETUP.md; for client engagements see docs/MULTI_CLIENT.md and the self-expiring onboarding stack.

No write action, anywhere. fix --write edits your local working copy so you can review it with git diff; it never calls a mutating AWS API.


What it will not do

Deliberate, and the reason to trust the numbers it does give:

  • Won't price from a public price list when it has never seen that usage type in your account. It says unpriced and explains why.

  • Won't claim a saving it can't replay against your real line items. Those say "estimate rather than a replay of history."

  • Won't report a resource as idle because it emits no activity data. Absence of evidence is not evidence of idleness.

  • Won't tell you to schedule something with no stopped state.

  • Won't reformat your Terraform to change one attribute.

  • Won't apply anything. You merge.

Confidence is stated on every figure — measured, derived or bounded — and never summed into one optimistic headline.


Status

253 tests. Full pipeline works end to end.

Validated against synthetic data only. The fixture plants findings with known values and computes ground truth independently of the code under test, so the logic and arithmetic are sound. What that cannot prove is that a real AWS bill matches the assumptions — the reconciliation tolerance (0.5%), the Terraform-type-to-usage-type mapping, and the flow-log field casing are all calibrated against data I wrote.

Running it against one real CUR is the highest-value next step, and the fastest way to find out which assumptions are wrong.

Available Tools

6 tools
capabilitiesA

What this server can and cannot do.

Worth asking before granting access: the answer is short because the tool is read-only by construction rather than by policy.

Inputs: no named parameters. Returns: result

free_tier_statusB

Free Tier allowances and how close this account is to each.

Not in the bill, so a charge sitting just over an allowance is invisible without it — and the saving there is the whole charge.

Inputs: profile. Returns: result

show_fixA

Generate the change for one finding, as a diff or a command.

Nothing is applied. The caller reviews and merges. Terraform edits are returned as a unified diff against the declaring file; resources not in code get the equivalent AWS CLI command instead.

Inputs: repo (required), finding_id (required), cur, profile. Returns: result

what_changedA

Explain a bill that moved: growth, the services driving it, and spikes.

Compares complete billing periods only. A month still running is not comparable to a finished one, and reporting it as a decline would be the calendar rather than the bill.

Inputs: cur, profile. Returns: result

what_is_wasting_moneyA

Find avoidable cloud spend on an AWS account.

Start here. Returns what the account costs, what looks avoidable, and for each finding the evidence, the fix, the blast radius and what breaks if the finding is wrong.

profile: AWS profile name. Uses the default credential chain if omitted. cur: Billing export location (s3://... ). Omit to use Cost Explorer, which needs no setup but gives no per-resource detail. repo: Local repo path, to resolve ownership and AI cost defects.

Inputs: cur, repo, profile. Returns: result

what_was_not_examinedA

Show which spend no rule looks at.

A report with three findings reads identically whether it inspected 95% of the bill or 5%. This is the denominator: every dollar is either flagged, checked and clean, or never examined. The unexamined list is ordered largest first.

Inputs: cur, profile. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool addresses a distinct aspect of cloud cost analysis: waste detection, coverage, changes, free tier, fix generation, and server capabilities. No two tools overlap in purpose.

    Naming Consistency3/5

    Tool names mix 'what_' question-style prefixes (what_is_wasting_money, what_was_not_examined, what_changed) with noun phrases (free_tier_status, capabilities) and one verb phrase (show_fix), creating inconsistent naming conventions.

    Tool Count5/5

    Six tools is well-scoped for a cost analysis server, covering the full workflow without redundancy or bloat.

    Completeness5/5

    The tool set provides a complete cost analysis lifecycle: identify waste, examine coverage, analyze changes, check free tier, and generate fixes. No obvious gaps within the read-only scope.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables natural language analysis of AWS cloud costs, waste audits, and budget insights across multiple profiles and regions.
      MIT
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables natural-language analysis of AWS cloud costs, waste and savings audits, and budget monitoring across multiple accounts and regions while keeping credentials local.
      -
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables users to analyze cloud costs, audit for waste, and get budget insights using natural language. It provides read-only access to AWS financial and resource data while keeping credentials on the local machine.
      -
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables users to analyze AWS costs, audit for waste, and get budget insights through natural language across AWS profiles and regions while keeping credentials local.
      -